INDIA
25 August 2015
From a PPP to Public Sector Control of Metronet in the UK: Implications for India
Mukul G. Asher, Shahana Sheikh and V. Ramakrishnan examine the implications for India of the restructuring of the UK's Metronet from PPP to full public sector control.
The primary objective of this article is to understand how a large transport sector PPP (Public Private Partnership) in UK, a high-income country, with a reputation for good public financial management, came to be restructured from a PPP to full public sector control in less than a decade, and the implications this episode holds for India, a middle-income country.
A critical examination of this episode is particularly relevant at this juncture as Indian government organizations, at both the Union and State government levels, hope to use the PPP method to complement traditional sources of public sector financing, undertaking construction and operations of infrastructure projects and public amenities.
The PPP under consideration is called the “Metronet PPP” which was carried out by the United Kingdom (UK) Government for the London Tube. This PPP, initiated in 2003, was designed to carry out refurbishment of a part of the London Underground. Envisioned as a 30 year project, the private holding company of the PPP was declared bankrupt in 2007, and was fully taken over by the public sector Transport for London (TfL) in 2008. Hence, the PPP was wound up a few years after its inception, with operations fully reverting back to the public sector organisation, and this continues until today.
It was envisaged that the public sector will gain from the PPP, as compared to conventional public procurement, due to stability of finance and the private sector’s project management skills. A 2002 report of the UK Parliament indicated that for the work under the PPPs, an investment of about GBP 16 billion (in terms of present value in 2002) was expected over the next 15 years; and as compared to conventional public sector funding, the PPP arrangements were expected to lead to a saving of GBP 2 billion and faster and more reliable journeys on the London Tube, as compared to the alternatives, worth as much as a further GBP 2 billion to the passengers through the life of the projects.
Under the PPP arrangements, the responsibility to carry out the renovation and refurbishment of different parts of the London Tube was handed over to private sector consortia under the three PPP contracts. However, the day-to-day functioning of the London Tube including responsibility for stations, train operations, signalling and safety, service patterns and setting fares was retained by the London Underground Limited (hereafter referred to as LUL), a public sector organisation.
The Department for Transport (DfT) at the UK Government would provide a yearly grant of GBP 1-1.1 billion to Transport for London (TfL) which would, in turn, give this to the three private consortia as payments for their services. The payments as annual “Infrastructure Service Charge” would vary based on four performance indicators. Further, the financing structure of the Metronet PPP was constituted by about 12 percent equity and the remaining as debt. The equity was split equally among five private companies which formed the private consortium for Metronet; while the debt was a combination of senior debt through bank loans (which would have to be paid off first in case of the failure of the PPP) and bonds. A clause in the PPP contracts required a review of the contracts at the interval of every 7.5 years, and also a decision at these intervals on the price of the contract for the next 7.5 year tranche. A PPP Arbiter’s post was created to mediate between the private consortia and LUL in case they were not able to agree on the price.
First, a significant part of Metronet’s obligations, nearly 60 percent of its projected capital expenditure in the first 7.5 year period, were to be delivered through contracts with its own five shareholders. This meant that each equity partner had contracts with Metronet to supply services for the work of refurbishment, implying a case of “tied supply chains”.
Tied supply chains have benefits but they require two pre-conditions: the scope of work is well established prior to the start of work, and a strong governance structure which while understanding the incentives of the shareholders, maintains tight oversight on the supply chain (Gannon et al. 2013). This implies that in a PPP arrangement, tied supply chain should be avoided, unless these two pre-conditions are satisfied – which was not the case in Metronet PPP, leading to a conflict of interest which remained unresolved and only profited the shareholders.
Second, the Metronet PPP case highlights too much dependence on the private sector for project management. According to the National audit Office (NAO) of UK, the project management arrangements actually remained unclear and a majority of the supply contracts were not competitively tendered under the Metronet PPP.
The main implication is that any government body or department, which is involved in a PPP or any such project, must get involved in the implementation with the aim of improving its own project management capabilities. The PPP Division (or Cell) at the Ministry of Finance, Government of India, is well-placed to improve project management abilities in the public sector. This could also enable retained in-house government knowledge (in areas including bid evaluation) to support further PPP engagements of government departments.
Third, given the financing structure of the Metronet PPP, it may be perceived that a bulk of the risk was with the debt owners. However, TfL had given a 95 percent guarantee to the senior debt providers, and the DfT had given an informal assurance to them that the financial debt obligations would be met. This created contingent fiscal liabilities for the UK Government, which turned into actual liabilities when the Metronet PPP failed. Further, the Metronet PPP had been classified as a “private non financial corporation” from 2003 to 2007 and hence, until then these liabilities were not accounted for in the national accounts.
The assurance by DfT meant that DfT actually faced the maximum amount of risk. However, it did not directly monitor the PPP and instead relied on other actors, who in turn did not act as expected. For instance, it relied on LUL to keep a track of the costs and performance of Metronet but LUL had access to only high-level cost information for Metronet and found it difficult to define various performance indicators; further, it relied on the shareholders of Metronet to monitor the costs too, but due to the tied supply chains, they had little incentive to reduce costs, instead they actually functioned as though they had agreed to cost-plus contracts, expecting to be paid for additional costs incurred.
This has two implications for PPPs in India. First, if any guarantees are given, they should be explicit and the related contingent liabilities, and the associated cost to the public sector in the case of a PPP failure, ought to be estimated at the inception stage. Second, the public agencies which face risks must have the necessary risk management processes in place. This also requires structures and human resource policies which enable government organisations to nurture needed skill-sets, much less rigid than the current practices.
Fourth, the Metronet faced two challenges with regard to its contract specifications. There was uncertainty regarding the information on assets and associated costs, and definitions of words such as “modernisation”, “refurbishment” and “enhanced refurbishment” were not clear, leaving them open for interpretation and consequently, repeated, time consuming disagreements between Metronet and LUL.
This implies that during the formulation of a PPP contract, processes of contract specification, including asset information and whole-life costing, need to be carried out in consultation with experts placed at a higher, more central level. In case of India, these efforts can be led by the PPP Division (or Cell) at the Ministry of Finance, Government of India. The Union Budget for 2014-15 had proposed the setting up of an entity called “3P India” which would be entrusted with the primary task of mainstreaming PPPs and increasing focus on delivery of efficient PPPs. Not only does such an entity merit consideration, but as NITI Aayog has recently suggested, it could carry out restructuring of existing PPP contracts due to specialised skills, which it is expected to house.
Moreover, sector specific units such as the PPP Division at the Indian Railways or National Highway Authority or the Ministry of Defence could also develop its own knowledge base; and can also review contracts to check for the language. In this context, the NITI Aayog has also suggested that every Ministry engaged in PPPs should create a dedicated division for monitoring of PPPs with full time staff and budgets to hire appropriate experts.
Fifth, cost and performance data on Metronet PPP was not adequately available and not regularly analysed. Indeed, assessment of efficiency requires relevant international benchmarking, but such exercises are time consuming, and require commitment from those involved, and hence, it is important to establish them early and maintain them to provide time series data. This implies that information and data systems for performance monitoring as well as for benchmarking would need specific attention. This would not only need to be established as soon as the PPP is implemented, but the data would also need to be periodically analysed to give feedback to the private and public sector stakeholders involved in the PPP, through the life of the project.
Sixth, the Metronet PPP contract and its close comparison with the similar Tube Lines PPP contract, reveals that as of 2007, in relative terms, the Tube Lines PPP was more successful. In other words, though the two PPPs were for the same purpose and similarly designed, the outcomes were different.
The above implies that it is not appropriate to rely on a “model PPP contract” or a “blueprint” for PPP projects even within the same sector. In the Indian context, it is necessary to note that standardised documents such as model concession agreements across infrastructure sectors were developed previously; however, concerns have been raised about their rigidity and the need to introduce greater flexibility for unforeseen circumstances. Risks and relationships among various stakeholders are unique in each PPP contract and hence, a “one-size fits all” approach is not appropriate. Further, since citizens’ needs and expectations change with time, every PPP contract must have a mechanism to identify and adapt the delivery framework, which implies that the framework has to be necessarily flexible, over its lifespan.
A remarkable feature of the Metronet PPP case is that when the failure emerged, the structures and institutions of the UK Government were able to reverse the PPP decision within a relatively short time of around five years, demonstrating welcome capacity for reversal, albeit at high cost. Indeed, it was on the horizon of the first 7.5 year periodic review that the cost escalations and other challenges facing the Metronet PPP started getting identified. This implies that, strong public sector institutions and periodic reviews have a key role to play in case of wrong decisions or failure of a PPP.
For the PPPs in India, weak public institutions have led the private sector to bear project implementation risks (and hence, associated costs) relating to activities such as delays in land acquisition and environmental clearances, and there has been an absence of structures for ex-ante negotiation. Both these would need urgent attention if indeed PPPs are to be used more widely.
Finally, the importance of requisite intent to be citizen-centric, and integrity of the system governing the relevant stakeholders, including public authorities and political leadership, cannot be overemphasized in the Indian context. A set of complementary reforms in government organizations, and better social and political norms consistent with broader public interest, are also needed for the PPP method to succeed. The strategic shift necessitates structural reform and systemic not parametric change.
A critical examination of this episode is particularly relevant at this juncture as Indian government organizations, at both the Union and State government levels, hope to use the PPP method to complement traditional sources of public sector financing, undertaking construction and operations of infrastructure projects and public amenities.
The PPP under consideration is called the “Metronet PPP” which was carried out by the United Kingdom (UK) Government for the London Tube. This PPP, initiated in 2003, was designed to carry out refurbishment of a part of the London Underground. Envisioned as a 30 year project, the private holding company of the PPP was declared bankrupt in 2007, and was fully taken over by the public sector Transport for London (TfL) in 2008. Hence, the PPP was wound up a few years after its inception, with operations fully reverting back to the public sector organisation, and this continues until today.
The Metronet PPP Arrangement
Before understanding the reasons for Metronet PPP’s failure, we first need to consider some details of the PPP arrangement. The Metronet PPP became operational in April 2003, and at the time it was the largest PPP in the UK. It was comprised two PPP contracts for renovation of two different parts of the London Tube: Metronet BCV (for the Bakerloo, Central and Victoria tube lines) and Metronet SSL (for the Sub Surface tube lines of the London Tube such as District, Circle, Metropolitan, Hammersmith & City, Waterloo & City tube lines). Alongside, another PPP contract, called the Tubelines PPP, was also operational for the remaining part of the London Tube (Jubilee, Northern, Piccadilly tube lines). Metronet and Tubelines were each a consortium of private sector companies.It was envisaged that the public sector will gain from the PPP, as compared to conventional public procurement, due to stability of finance and the private sector’s project management skills. A 2002 report of the UK Parliament indicated that for the work under the PPPs, an investment of about GBP 16 billion (in terms of present value in 2002) was expected over the next 15 years; and as compared to conventional public sector funding, the PPP arrangements were expected to lead to a saving of GBP 2 billion and faster and more reliable journeys on the London Tube, as compared to the alternatives, worth as much as a further GBP 2 billion to the passengers through the life of the projects.
Under the PPP arrangements, the responsibility to carry out the renovation and refurbishment of different parts of the London Tube was handed over to private sector consortia under the three PPP contracts. However, the day-to-day functioning of the London Tube including responsibility for stations, train operations, signalling and safety, service patterns and setting fares was retained by the London Underground Limited (hereafter referred to as LUL), a public sector organisation.
The Department for Transport (DfT) at the UK Government would provide a yearly grant of GBP 1-1.1 billion to Transport for London (TfL) which would, in turn, give this to the three private consortia as payments for their services. The payments as annual “Infrastructure Service Charge” would vary based on four performance indicators. Further, the financing structure of the Metronet PPP was constituted by about 12 percent equity and the remaining as debt. The equity was split equally among five private companies which formed the private consortium for Metronet; while the debt was a combination of senior debt through bank loans (which would have to be paid off first in case of the failure of the PPP) and bonds. A clause in the PPP contracts required a review of the contracts at the interval of every 7.5 years, and also a decision at these intervals on the price of the contract for the next 7.5 year tranche. A PPP Arbiter’s post was created to mediate between the private consortia and LUL in case they were not able to agree on the price.
Implications for India
The main question discussed is what challenges led to the Metronet PPP being transferred to public ownership and control, and what are the policy and operational implications of this episode for pursing the PPP method in India.First, a significant part of Metronet’s obligations, nearly 60 percent of its projected capital expenditure in the first 7.5 year period, were to be delivered through contracts with its own five shareholders. This meant that each equity partner had contracts with Metronet to supply services for the work of refurbishment, implying a case of “tied supply chains”.
Tied supply chains have benefits but they require two pre-conditions: the scope of work is well established prior to the start of work, and a strong governance structure which while understanding the incentives of the shareholders, maintains tight oversight on the supply chain (Gannon et al. 2013). This implies that in a PPP arrangement, tied supply chain should be avoided, unless these two pre-conditions are satisfied – which was not the case in Metronet PPP, leading to a conflict of interest which remained unresolved and only profited the shareholders.
Second, the Metronet PPP case highlights too much dependence on the private sector for project management. According to the National audit Office (NAO) of UK, the project management arrangements actually remained unclear and a majority of the supply contracts were not competitively tendered under the Metronet PPP.
The main implication is that any government body or department, which is involved in a PPP or any such project, must get involved in the implementation with the aim of improving its own project management capabilities. The PPP Division (or Cell) at the Ministry of Finance, Government of India, is well-placed to improve project management abilities in the public sector. This could also enable retained in-house government knowledge (in areas including bid evaluation) to support further PPP engagements of government departments.
Third, given the financing structure of the Metronet PPP, it may be perceived that a bulk of the risk was with the debt owners. However, TfL had given a 95 percent guarantee to the senior debt providers, and the DfT had given an informal assurance to them that the financial debt obligations would be met. This created contingent fiscal liabilities for the UK Government, which turned into actual liabilities when the Metronet PPP failed. Further, the Metronet PPP had been classified as a “private non financial corporation” from 2003 to 2007 and hence, until then these liabilities were not accounted for in the national accounts.
The assurance by DfT meant that DfT actually faced the maximum amount of risk. However, it did not directly monitor the PPP and instead relied on other actors, who in turn did not act as expected. For instance, it relied on LUL to keep a track of the costs and performance of Metronet but LUL had access to only high-level cost information for Metronet and found it difficult to define various performance indicators; further, it relied on the shareholders of Metronet to monitor the costs too, but due to the tied supply chains, they had little incentive to reduce costs, instead they actually functioned as though they had agreed to cost-plus contracts, expecting to be paid for additional costs incurred.
This has two implications for PPPs in India. First, if any guarantees are given, they should be explicit and the related contingent liabilities, and the associated cost to the public sector in the case of a PPP failure, ought to be estimated at the inception stage. Second, the public agencies which face risks must have the necessary risk management processes in place. This also requires structures and human resource policies which enable government organisations to nurture needed skill-sets, much less rigid than the current practices.
Fourth, the Metronet faced two challenges with regard to its contract specifications. There was uncertainty regarding the information on assets and associated costs, and definitions of words such as “modernisation”, “refurbishment” and “enhanced refurbishment” were not clear, leaving them open for interpretation and consequently, repeated, time consuming disagreements between Metronet and LUL.
This implies that during the formulation of a PPP contract, processes of contract specification, including asset information and whole-life costing, need to be carried out in consultation with experts placed at a higher, more central level. In case of India, these efforts can be led by the PPP Division (or Cell) at the Ministry of Finance, Government of India. The Union Budget for 2014-15 had proposed the setting up of an entity called “3P India” which would be entrusted with the primary task of mainstreaming PPPs and increasing focus on delivery of efficient PPPs. Not only does such an entity merit consideration, but as NITI Aayog has recently suggested, it could carry out restructuring of existing PPP contracts due to specialised skills, which it is expected to house.
Moreover, sector specific units such as the PPP Division at the Indian Railways or National Highway Authority or the Ministry of Defence could also develop its own knowledge base; and can also review contracts to check for the language. In this context, the NITI Aayog has also suggested that every Ministry engaged in PPPs should create a dedicated division for monitoring of PPPs with full time staff and budgets to hire appropriate experts.
Fifth, cost and performance data on Metronet PPP was not adequately available and not regularly analysed. Indeed, assessment of efficiency requires relevant international benchmarking, but such exercises are time consuming, and require commitment from those involved, and hence, it is important to establish them early and maintain them to provide time series data. This implies that information and data systems for performance monitoring as well as for benchmarking would need specific attention. This would not only need to be established as soon as the PPP is implemented, but the data would also need to be periodically analysed to give feedback to the private and public sector stakeholders involved in the PPP, through the life of the project.
Sixth, the Metronet PPP contract and its close comparison with the similar Tube Lines PPP contract, reveals that as of 2007, in relative terms, the Tube Lines PPP was more successful. In other words, though the two PPPs were for the same purpose and similarly designed, the outcomes were different.
The above implies that it is not appropriate to rely on a “model PPP contract” or a “blueprint” for PPP projects even within the same sector. In the Indian context, it is necessary to note that standardised documents such as model concession agreements across infrastructure sectors were developed previously; however, concerns have been raised about their rigidity and the need to introduce greater flexibility for unforeseen circumstances. Risks and relationships among various stakeholders are unique in each PPP contract and hence, a “one-size fits all” approach is not appropriate. Further, since citizens’ needs and expectations change with time, every PPP contract must have a mechanism to identify and adapt the delivery framework, which implies that the framework has to be necessarily flexible, over its lifespan.
A remarkable feature of the Metronet PPP case is that when the failure emerged, the structures and institutions of the UK Government were able to reverse the PPP decision within a relatively short time of around five years, demonstrating welcome capacity for reversal, albeit at high cost. Indeed, it was on the horizon of the first 7.5 year periodic review that the cost escalations and other challenges facing the Metronet PPP started getting identified. This implies that, strong public sector institutions and periodic reviews have a key role to play in case of wrong decisions or failure of a PPP.
For the PPPs in India, weak public institutions have led the private sector to bear project implementation risks (and hence, associated costs) relating to activities such as delays in land acquisition and environmental clearances, and there has been an absence of structures for ex-ante negotiation. Both these would need urgent attention if indeed PPPs are to be used more widely.
Finally, the importance of requisite intent to be citizen-centric, and integrity of the system governing the relevant stakeholders, including public authorities and political leadership, cannot be overemphasized in the Indian context. A set of complementary reforms in government organizations, and better social and political norms consistent with broader public interest, are also needed for the PPP method to succeed. The strategic shift necessitates structural reform and systemic not parametric change.