LE A R N ING PR O DUC T Public-Private Partnerships in Health World Bank Group Engagement in Health PPPs An IEG Synthesis Report © 2016 International Bank for Reconstruction This work is a product of the staff of The World RIGHTS AND PERMISSIONS and Development / The World Bank Bank with external contributions. The findings, The material in this work is subject to copyright. 1818 H Street NW interpretations, and conclusions expressed in Because The World Bank encourages Washington, DC 20433 this work do not necessarily reflect the views dissemination of its knowledge, this work Telephone: 202-473-1000 of The World Bank, its Board of Executive may be reproduced, in whole or in part, for Internet: www.worldbank.org Directors, or the governments they represent. noncommercial purposes as long as full attribution to this work is given. The World Bank does not guarantee the accuracy of the data included in this work. Any queries on rights and licenses, including The boundaries, colors, denominations, and subsidiary rights, should be addressed to other information shown on any map in this World Bank Publications, The World Bank work do not imply any judgment on the part Group, 1818 H Street NW, Washington, DC of The World Bank concerning the legal 20433, USA; fax: 202-522-2625; e-mail: status of any territory or the endorsement pubrights@worldbank.org. or acceptance of such boundaries. Public-Private Partnerships in Health World Bank Group Engagement in Health PPPs An IEG Synthesis Report Contents ACKNOWLEDGMENTS ........................................................................................................................ IV ABBREVIATIONS .................................................................................................................................. V OVERVIEW............................................................................................................................................ VI MANAGEMENT COMMENTS ................................................................................................................ X 1. BACKGROUND ......................................................................................................................... 1 Defining PPPs in the Health Sector .......................................................................................................................... 3 Objectives ................................................................................................................................................................. 4 Scope and Methodology ........................................................................................................................................... 5 2. WORLD BANK GROUP PORTFOLIO ..................................................................................... 7 3. RELEVANCE OF WORLD BANK GROUP SUPPORT ........................................................... 10 4. COORDINATION ACROSS THE WORLD BANK GROUP ..................................................... 15 5. IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE ................................. 18 Lessons from the Structuring Process ......................................................................................................................................... 18 PPPs and the Poor ....................................................................................................................................................................... 22 Lessons from Operating PPPs ..................................................................................................................................................... 23 6. LESSONS LEARNED.............................................................................................................. 27 iii Acknowledgments The report was prepared by a team led by Stefan Apfalter (team leader and senior evaluation officer, IEGFP) and comprising Maria Elena Pinglo (evaluation officer, IEGSD), Antonio Giuffrida (lead evaluation specialist, IEGHC), Victor Malca, and Houqi Hong (consultants, IEGFP). The review was conducted under the supervision of Marvin Taylor-Dormond (Director, IEGSP), Stoyan Tenev (senior manager, IEGFP) and Marie Gaarder (manager, IEGHC). Emelda Cudilla provided program support. The study also benefited from background research conducted by Salvatore Schiavo-Campo (consultant, IEG). The team gratefully acknowledges the contributions of present and former World Bank Group managers and staff who assisted in the report. Evaluation Managers  Caroline Heider Director-General, Evaluation  Marvin Taylor-Dormond Director, Financial, Private Sector, and Sustainable Development Department  Stoyan Tenev Senior Manager, Financial and Private Sector Development Unit  Marie Gaarder Manager, Human Development and Corporate Program Unit  Stefan Apfalter Task Team Leader, Financial and Private Sector Development Unit iv Abbreviations AS Advisory Services CAS Country Assistance Strategy CPF Country Partnership Frameworks DBFO design, build, finance, and operate EIU Economist Intelligence Unit HNP Health, Nutrition, and Population HSR health sector reform ICR Implementation Completion Report IDA International Development Association IEG Independent Evaluation Group IFC International Finance Corporation M&E monitoring and evaluation PCR Project Completion Report PFI Private Finance Initiative PFRAM PPP Fiscal Risk Assessment Model PPAR Project Performance Assessment Report PPP public-private partnership PSD private sector development SEA strategic assessment area SDG Sustainable Development Goal UHC universal health coverage v Overview Access to essential health services is an important aspect of development. Governments from both developed and developing countries are increasingly looking at public-private partnerships (PPPs) as a way to expand access to higher-quality health services by leveraging capital, managerial capacity, and knowhow from the private sector. Originally confined to the traditional infrastructure sectors of transport, water, or energy, PPPs are increasingly applied also in social infrastructure sectors, particularly for delivery of health services. PPPs and other forms of private sector involvement in health are now also an important element of the World Bank Group’s response to country health challenges, as reflected in the 2013 World Bank Group Strategy, the 2008 World Bank Group Health Development Strategy, the 2015 joint World Bank Group Approach to Harnessing the Private Sector in Health, various CASs and CPFs and IFC’s FY17-19 and prior Strategy and Business Outlook reports. Such a complex form of contracting is likely to generate distinctive set of challenges in developing country settings. Entering into PPP arrangements imposes certain requirements on the enabling environment of the respective country and bears risks; however, the same also applies to the alternatives of remaining inactive or continuing service delivery through the public sector. The objective of this review is to provide insights into the Bank Group’s work of applying PPP arrangements in the health sector, to distill knowledge with regard to what works (and what does not), review the quality of work in structuring PPP arrangements, and identify lessons to be learned from successful and failed efforts to structure health PPPs approved during FY04-15. The review encompasses all institutions of the Bank Group engaged in PPPs in health. The approach is based on a desk-based portfolio review of Bank Group projects, relying on existing evaluative evidence, primarily project-level evaluation reports and data. Such an approach has its limitations: project-level evaluation reports are prepared at a quite early stage in PPP contracts whose duration often ranges from 5–30 years. They mainly shed light on the structuring process, but offer little or no evidence as to what happened beyond commercial closure of the PPP. Second, this review has been commissioned at a time when only a few IFC investments in PPPs have been evaluated. Only one out of four investments have reached operational maturity and, hence, are being evaluated; therefore, conclusions on the operational effectiveness of PPPs and their impacts are not possible. The findings in this review have to be seen in this context, that is, as an assessment using a narrowly defined concept of success, without the ambition to pass a judgment on the longevity of PPPs or on their effects on shared prosperity or reducing poverty. KEY INSIGHTS From FY04 to FY15, the Bank Group approved 78 projects that provided support for health-related PPP operations, of which 53 were IFC projects (49 of these are advisory projects, and four are IFC investments) and 25 World Bank lending projects. More projects (58 percent) focused on downstream issues (structuring and finance) than on upstream issues (42 percent) of policy and sectoral advice. vi OVERVIEW Capacity building was an increasingly prominent focus of the World Bank’s upstream support. This represents a much-needed effort to strengthen countries’ ability to assess the merits of deploying PPPs in the health sector and their knowledge of how to structure such engagements. The 2013 IEG evaluation, World Bank Group Support to PPPs—Lessons from Experience in Client Countries FY2012, identified the lack of local skills and resources for the preparation of a PPP pipeline to be a serious limitation across most Bank Group–supported countries. IFC Advisory Services (AS) focused on downstream support by providing structuring advice and finance. Bank Group support exhibited significant variability over time. The Bank Group supported countries that need its support the most. More specifically, the World Bank engages mostly in countries where the enabling environment for PPPs is weak, and where the legal and regulatory frameworks have not yet evolved, that is, in “nascent countries” according to a classification scheme of the Economist Intelligence Unit (EIU). On the other hand, IFC supports more PPPs in countries that the EIU classifies as having “emerging” and “mature” PPP environments, helping them to develop and deepen the PPP market through advice on structuring PPPs and investments for specific opportunities. Furthermore, Bank Group support for health PPPs was relevant to countries in as much as it supported clear development priorities, that is, insofar as the PPPs addressed issues that were high on the agenda of the country development strategy as expressed by the Country Assistance Strategy or other related strategy documents. Overall, the analysis indicates that the type of health PPPs supported by the Bank Group tend to be more short-term arrangements, compared to infrastructure PPPs. Health PPPs can exhibit considerable complexity, particularly in the clinical operation of such facilities. Some types of health PPP can be of lesser complexity with shorter duration, and because they typically involve contracts for the delivery of quite narrowly defined health services, they are at the lower end of the risk spectrum. The most commonly supported PPP is combined health service provision, which includes construction or refurbishment, operation and provision of clinical services. Although this type of PPP seems the most complex, there is significant variation within the category, placing different projects at different points along the risk spectrum, ranging from, for example, a 10-year concession for the construction and provision of a narrowly defined health service unit (for example, diagnostic and imaging unit) to a 20-year concession for the construction and operation of a hospital and clinics with a broad range of health services provision (as in Lesotho). However, the review found that this latter type are rare. The analysis of Bank Group interventions along the entire value chain from policy to financing shows that adequate sequencing is rare. Sequencing Bank Group interventions would prepare client countries for private sector involvement through regulatory or legal advice or defining the space where such involvement would be desirable, before actual PPPs are structured or financed. In all but two countries, the Bank provided Health, Nutrition, and Population (HNP) lending toward general health sector reforms and to fight specific diseases or health conditions (other type of HNP support). In only a few cases, however, do countries appear to have received timely private sector development (PSD) support for health or specific support toward health PPPs. vii Of all IFC AS health mandates, 63 percent reached contract closure. Among projects that led to contract closure, the key success factors are government commitment and sound technical design. Bank Group advice has not been strategic enough with respect to laying out all options for the provision of health services in a given country. IEG reviewed the extent to which IFC AS presented public procurement as an option; of 12 Strategic Options Reports reviewed by IEG, public procurement was presented in one case. The demand-driven nature of IFC AS may explain this finding. However, the efficiency and desirability from a social perspective of the PPP cannot be established without a comparison with the alternatives, the main one being the public option. This needs to be done ex-ante, during, and ex-post the PPP transaction. There is little evidence that fiscal implications are assessed consistently, even if the proposed PPP could have significant fiscal implications. This largely corroborates the findings of the 2013 PPP evaluation. Following the recommendation of that 2013 evaluation, the Bank Group, together with the International Monetary Fund (IMF), developed a PPP Fiscal Risk Assessment Model (PFRAM).1 It will be essential to apply this tool kit systematically to PPPs that could affect the fiscal space of client governments. Self-assessment reports (prepared two years after completion) indicate positive effects of PPPs in some areas, for example, access and quality; but evidence is still limited. The current Bank Group monitoring and evaluation (M&E) system for health PPPs is inadequate and needs to be improved to better track results. Currently, insufficient data limits learning and poses a reputational risk for the Bank Group. The review found that the vast majority of World Bank Group–supported PPP interventions have an explicit emphasis on the poor or underserved population in their design. However, suitable M&E indicators or proper baselines and targets—needed to track results and assess whether the poor were able to access the health services—are lacking. The 2014 OECD report2 indicates that although post-contract award management of PPP projects is indeed critical for securing the expected benefits, it has generally not been an important priority for governments. Despite the need, the Bank Group rarely provides “aftercare” for contract management. Providing aftercare would also give the Bank Group an opportunity to learn ‘first hand’ about post-financial closure results. As the development community and the Bank Group increasingly partner with the private sector to improve the delivery of health services, this review finds five lessons: (1) When advising governments on the various models for providing health services, both the IFC and the World Bank should act as one Bank Group and discuss the whole range of options from the public and mixed options to the various types of potential health PPPs, taking account of the sector reform context of a country and 1PFRAM is available at https://www.imf.org/external/np/fad/publicinvestment/pdf/PFRAM.pdf 2 Guasch, J. L. et al. (2014), “The Renegotiation of PPP Contracts: An Overview of Its Recent Evolution in Latin America”, International Transport Forum Discussion Papers, 2014/18, OECD Publishing, Paris. http://dx.doi.org/10.1787/5jrw2xxlks8v-en viii OVERVIEW its overall maturity and track record in using PPPs. (2) To be better positioned to deliver such strategic advice, the Bank Group should better integrate its sector reform and policy work on the structuring and financing of PPPs. (3) Pro-poor access and affordability need not only be systematically considered at the design stage; they should also be tracked to ensure that the poor actually benefit from PPPs. (4) The PFRAM tool should be systematically applied to the structuring of PPPs with substantial fiscal implications.3 (5) Given the long contractual life of PPPs, the new practice of preparing post- completion reports after the PPPs have gone into operation should be mainstreamed so that these reports can be conducted at an appropriate time. 3 PPP projects risks can be divided into five somewhat overlapping categories: construction risk, financial risk, availability risk, demand risk, and residual value risk (see IMF, 2006. Public-Private Partnerships, Government Guarantees and Fiscal Risk. IMF, Washington DC). ix Management Comments INTRODUCTION The World Bank Group (WBG) thanks IEG for the Synthesis Report on WBG’s support to public- private partnerships (PPPs) in health. Access to affordable, quality health care is crucial for people’s well-being and is a key driver of economic growth and development. Governments play a central role in the provision and regulation of health care. However, there is an increasing recognition that, on their own, governments cannot deliver enough services to achieve universal health coverage and meet their populations’ needs. All players in the health area, including the private sector, will need to be involved if countries are to deliver universal health coverage and meet SDG 3 on Health. PPPs are not a panacea, but when well designed, they can strengthen the quality and efficiency of public health services. Well-structured PPPs can create clearly defined platforms that involve the private sector in delivering solutions to health challenges in developing countries. They can also help countries improve their ability to regulate the private sector and find an efficient and effective mix of public and private sector participation in the health sector. The Management of the WBG institutions broadly agrees with the findings of the IEG report, which it finds balanced. These comments describes how IEG’s lessons are being implemented. WORLD BANK GROUP COMMENTS ON LESSONS Strategic importance of health PPPs and alignment with country needs. Management welcomes the IEG finding that WBG support for health PPPs was relevant to countries and that these PPPs supported clear development priorities that were an important part of country development strategies. The report articulates the complementary roles of the WBG institutions within the PPP delivery chain, and captures their specific contributions. As the report points out, WBG support to health PPPs works across the health care system, from upstream support for the enabling environment, capacity building, and pipeline development, to downstream transactions and execution. Consideration of a range of public, private, and mixed options for addressing health system challenges. Management agrees that governments should consider the full range of options to find the best possible solution for resolving health system challenges. When developing Country Partnership Frameworks and health sector program strategies, the WBG considers a variety of approaches for increasing access to and improving the quality of health services. IFC Advisory Services becomes engaged where there is a clear and identified need to strengthen a public health service that a government has struggled to maintain, and where a PPP has been identified as an appropriate solution. Before structuring the PPP, IFC Advisory Services systematically conducts a technical and financial assessment of the way public services are delivered to identify the x potential gains or losses from private sector participation and to determine whether a PPP is an appropriate option to pursue. Sequencing of WBG interventions. Management agrees with IEG’s finding that proper sequencing of WBG interventions could better prepare countries for private sector involvement. As IEG highlighted, the World Bank—and many other development partners—mostly supports upstream activities such as capacity building, sector reform, and legal and regulatory frameworks, while PPP transactions are typically supported in countries where the enabling environment is ready for their implementation. The WBG has paid increasing attention to helping countries build their capacity to manage and implement PPPs. From 2002 to 2012, the WBG implemented nearly 800 capacity-building activities to help governments gain the skills and knowledge to develop and manage PPPs. The WBG has also developed a Country PPP Readiness Diagnostic Tool, which is being piloted and refined and will be more widely deployed in future WBG-supported PPP projects. World Bank Group coordination. The WBG fully agrees that a Group-wide coordinated approach to health sector strengthening is essential to support lasting and sustainable improvements in health care provision and management. Every country’s context and needs are unique, so PPPs are not always the optimum solution. When PPPs are identified as a possible solution for a country, the WBG is committed to improving coordination among its institutions in assessing the desirability and feasibility of PPPs and working collaboratively to ensure that any PPP solution is embedded in the government’s larger health sector policies and contributes to strengthening the public health system. In the past year the WBG has reviewed its processes, and IFC and HNP GP are putting in place mechanisms to ensure improved coordination. Access and affordability for the poor. Management welcomes the IEG finding that the designs of most World Bank-supported PPP interventions have an explicit emphasis on poor or underserved populations, and that the majority of health PPPs supported by IFC Advisory Services have a poverty focus. This is a welcome recognition of WBG efforts to use PPPs to expand health benefits, including to people in IDA countries and low-income states in middle-income countries. Management agrees on the need for project documents to be more explicit about how poor populations are expected to benefit from a health PPP. As part of the post-completion monitoring of its health PPP projects, IFC Advisory Services will aim to track access for poor populations wherever reliable data are available and accessible to the Bank Group. Concerted effort to consider fiscal impacts of health PPPs. Management agrees with IEG about the importance of assessing the fiscal implications of health PPPs for governments, especially given PPPs’ long duration. IFC’s Advisory Services conducts a fiscal assessment as part of every PPP advisory project. However, Management acknowledges that for past projects there could have been a more systematic approach to documenting these assessments, even when the fiscal impact was positive or neutral to the government. Going forward, the WBG will mandate the recording of these assessments in the project documents for health PPPs. Management also agrees that the fiscal assessments can be more systematic. Since the IEG Evaluation of WBG PPPs released in 2014, the PPP CCSA has worked with the IMF in developing the PPP Fiscal Risk Assessment Model (PFRAM) to support the fiscal assessment of PPP projects. Management agrees with the IEG lesson to use the PFRAM to assess the fiscal implications of all future health PPPs that carry substantial fiscal implications. xi Post-transaction monitoring of health PPPs. Management appreciates the lesson on improving the monitoring frameworks of health PPPs during the post-completion phase. In the Management Response to the 2014 IEG Evaluation of WBG PPPs, IFC Advisory Services committed to implement a program of post-completion monitoring for all successfully closed projects. Management welcomes IEG’s recognition that this step has been implemented since 2013, and the program will continue. The PPP CCSA has also taken steps to implement IEG’s 2014 recommendation and has conducted a review of the post-monitoring status and options for PPPs across the WBG. Post-transaction support for health PPPs. Management fully agrees with the IEG lesson on the need for post-contract management support for client governments that lack capacity and identify the need for it. Management appreciates IEG’s recognition that IFC Advisory Services’ post-transaction advisory support service, which was established in 2013, provides this kind of aftercare. IFC will continue to proactively assess client capability and needs to implement PPPs, and will offer post-transaction advisory support whenever it is needed. xii 1. Background Access to essential health services is an important aspect of development. One of the targets for the third Sustainable Development Goal ensuring healthy lives and promoting well-being for all is to achieve universal health coverage, including financial risk protection, access to quality essential health care services, and access to safe, effective, and affordable essential medicines and vaccines for all. Governments under increasing budgetary pressure are looking to public-private partnerships (PPPs) as a way to expand access to higher quality health services, leveraging capital, managerial capacity, and know-how from the private sector. The underlying causes of this increased interest in PPPs are the rising costs of delivering health care services as population age in developed and developing countries; increases in chronic diseases; and rapidly changing and advancing medical technologies. Originally confined to the traditional infrastructure sectors of transport, water, or energy, PPPs are also increasingly being applied in the social infrastructure sectors, particularly in the delivery of health services. Although there is a substantial body of research on the outcomes from PPPs in high-income countries,4,5 we know very little about applying the concept of PPPs in low- and middle-income countries and even less about outcomes. Therefore, little guidance is available to support Bank Group task team leaders (TTLs) and external policy makers in deciding whether and when private sector involvement in the delivery of health services in the form of PPPs is likely to be beneficial. Limited knowledge exists also with regard to the implementation challenges arising from structuring PPPs and managing the associated service contracts. Given the distinctive set of challenges that such a complex form of contracting is likely to generate in developing-country settings, these are important gaps that this evaluation synthesis would like to contribute to addressing. The Bank Group in its strategy is committed to promoting PPPs because “…such partnerships can contribute to improved basic service provision in areas—such as health […]—that are 6 essential for reducing poverty and boosting shared prosperity.” PPPs and other forms of private sector involvement in health are also important elements of other Bank Group strategies, including the 2008 World Bank Group Health Development Strategy, various Country Assistance Strategies (CASs) and Country Partnership Frameworks (CPFs), and IFC’s FY17-19 and prior Strategy and Business Outlook reports. The 2015 joint World Bank Group Approach to Harnessing the Private Sector in Health focuses on an integrated health system approach, regardless of whether it is public or private, and it recognizes that universal health care cannot be achieved without the private sector. 4 European Union, 2013. Health and Economics Analysis for an Evaluation of the Public-Private Partnerships in Health Care Delivery across the EU. 5 James Barlow, Jens Roehrich, and Steve Wright. “Europe Sees Mixed Results from Public -Private Partnerships for Building and Managing Health Care Facilities and Services. Health Affairs 32, no.1 (2013):146–154. 6 See The World Bank, 2013. World Bank Group Strategy (page 20). 1 CHAPTER 1 BACKGROUND The proposed LP will, therefore, assess the performance of a potentially important, albeit not yet widely used instrument that has gained important traction in the global health debate.7 The proposed study is closely aligned with the IEG strategy of focusing on the World Bank’s effectiveness in addressing clients’ development challenges and contributing to the Twin Goals of shared prosperity and reducing poverty. Thematically, the proposed product is largely in the IEG Strategic Engagement Area (SEA) of Sustained Service Delivery to the Poor. As such, it will contribute to the rising body of knowledge in this SEA and to the planned FY18 evaluation on improving access to essential health care services. This synthesis paper builds on previous IEG work on PPPs and on private sector involvement in health, notably the 2009 Health, Nutrition, and Population Evaluation8 and the 2013 World Bank Group Support to Public-Private Partnerships: Lessons from Experience in Client Countries FY02– 12. The 2009 Health, Nutrition, and Population Evaluation was intended to inform the implementation of HNP strategies so as to make future support more effective. Although the HNP Evaluation was broader in scope some of its findings were relevant to World Bank and IFC health, and echoed the findings of the 2013 PPP evaluation. For example, to strengthen the Bank Group’s ability to help countries to improve the efficiency of health systems, the IFC should: (i) support PPPs through Advisory Services (AS) to government and industry and through its investments, and expand investments in health insurance; and (ii) improve collaboration and joint sector work across the Bank Group, leveraging World Bank sector dialogue on health regulatory frameworks to engage new private actors and more systematically coordinate with the World Bank’s policy interventions in private sector participation in health. The more recent 2014 IEG review of Bank Group support to health financing over FY03–12,9 found that 70 percent of all IFC AS in health was dedicated to PPP projects, and highlighted the role of the Bank Group’s Health in Africa Initiative in fostering private sector participation in publicly funded health.10 The 2013 global evaluation of Bank Group support to PPPs revealed broad-based challenges applicable to the instrument of PPPs, even though the evaluation builds upon a review of a Bank Group portfolio comprising mostly infrastructure projects (Box 1.1). This report recommended to Bank Group Management: i) Translate the Bank Group’s strategic PPP intentions into an operational framework; ii) Better assist governments in making strategic decisions regarding the level and nature of private sector participation, and in assessing fiscal implications; iii) Identify avenues to increase IFC investments in PPPs in countries and markets that do not yet have a well-developed enabling environment; iv) Ensure broad stakeholder consultation and government commitment in IFC’s advisory work; v) Provide authoritative guidance to staff on how to handle unsolicited PPP proposals; and vi) Define principles for monitoring PPPs over the long term so as to capture all vital performance aspects of PPPs, including—where relevant—user aspects. Since 2013 IEG has observed substantial efforts in implementing these recommendations: for example, the recent adoption of the Public-Private Partnership Fiscal Risk Assessment Model (PFRAM), jointly created by the World Bank and the International Monetary Fund (IMF), for assessing fiscal implications. 7 WHO. 2015c. Trade, Foreign Policy, Diplomacy and Health: Public –Private Partnerships for Health. http://www.who.int/trade/glossary/story077/en/ (accessed June 20, 2015). 8 http://ieg.worldbank.org/Data/reports/hnp_full_eval.pdf 9 http://ieg.worldbank.org/evaluations/wbg-support-health-financing 10 https://www.wbginvestmentclimate.org/advisory-services/health/health-in-africa/ 2 CHAPTER 1 BACKGROUND This review provides an opportunity to assess the relevance of these previous findings and the recommendations of the 2009 and 2013 evaluations in the light of challenges posed by the health sector. Box 1.1. Findings from IEG’s Previous Evaluations on PPPs The 2013 evaluation found that (i) sector reform is crucial to PPP success, but such efforts often fail because of their inherent complexity and because of the political implications they typically entail; (ii) the Bank Group rarely gave advice on how to assess and manage fiscal implications of PPPs; (iii) the lack of local skills and resources for the preparation of a PPP pipeline and bankable PPP projects poses a serious limitation in most World Bank– supported countries; (iv) PPPs supported by the Bank Group were largely rated successful according to their immediate development outcome rating, but data are scarce on their actual performance, including fiscal implications and effects on the poor; (v) With regard to IFC Advisory Services, IEG found that volatile government commitment is the single most important factor in the failure of PPP structuring projects; (vi) Although the three Bank Group institutions deploy their respective comparative advantages well, their approach should be more strategic and better tailored to countries. Source: IEG 2013 “World Bank Group Support to Public-Private Partnerships: Lessons from Experience in Client Countries FY02–12,” http://ieg.worldbank.org/Data/reports/hnp_full_eval.pdf Defining PPPs in the Health Sector This review adopts a broad definition of PPP, in line with Bank Group policies. Accordingly, PPPs are “a long-term contract between a private party and a government entity, for providing a public asset or service, in which the private party bears significant risk and management responsibility, and remuneration is linked to performance.”11 PPPs can leverage capital, managerial capacity, and knowhow from the private sector. PPPs encompass a wide variety of arrangements. For example, the initial injection of capital by the private partners is a key component of many PPPs, particularly those in the hospital sector.12 However, PPPs in the healthcare sector vary in the scope of services covered. The external literature suggests the typology summarized in Table 1.1 below, which IEG adopted for this study. The Bank Group’s interventions tend to use all of these PPP models.13 A common critical element of the different types of PPPs is that the private party shares risks with the government. However, the degree of risk sharing between the private and public partners is not homogenous; it depends upon the level of capital committed by the private 11 The World Bank, Asian Development Bank, and Inter-American Development Bank 2014. Public- Private Partnerships Reference Guide. Version 2.0. 12 Martin McKee, Nigel Edwards, Rifat Atun. Public –Private Partnerships for Hospitals. Bulletin of the World Health Organization 2006; 84:890–896. 13 Alternative typologies available in the health PPP literature identify the full health services provision model, where a private operator builds and operates a hospital and some or all associated community primary care provision, with a contract to provide care for a defined geographic area; and co-location, where a public agency allocates a portion of a public hospital’s land or premises for use by a private operator in exchange for payment and specified benefits to the public agency. 3 CHAPTER 1 BACKGROUND party, the length of the partnership, the provision for renegotiation, and how payment mechanisms are structured. Table 1.1. A Typology of PPPs in the Healthcare Sector PPP model Common term Definition / Explanation Health services only Operating contract, A private operator is brought in to operate and deliver (selective) performance-based contract publicly funded health services in a publicly owned (concession, lease) facility. Facility finance Design, build, finance, A public agency contracts a private operator to (accommodation operate(DBFO), build, own, design, build, finance, and operate a hospital facility. only) operate, transfer (BOOT), Health services within the facility are (mostly) UK’s PFI provided by government. Combined Twin accommodation/ A private operator builds or leases a facility and (accommodation and clinical services joint provides free (or subsidized) healthcare services to a health services) venture/ Franchising, PFI+ defined population. Source: Adapted from Montagu and Harding (2012) and Barlow, Roehrich, and Wright (2013). Note: PFI = private finance initiative. Objectives The objective of this review is to provide insights into the Bank Group’s work on applying PPP arrangements in the health sector, to distill knowledge of what works (and what does not), to review the quality of the work in structuring PPP arrangements, and to identify lessons to be learned from successful and failed efforts to structure health PPPs. More specifically, the study seeks to answer the following questions: 1. Which products, tools, and services does the Bank Group deploy to support client countries in applying the PPP concept in the health sector? And how has its support evolved over time? 2. What can we learn from the way the Bank Group advises client governments on whether and how to use PPPs in health? a. How did the Bank Group support the conceptualization of PPP in the health sector (identification stage) and what can we learn from this experience? b. How did the Bank Group ensure that PPP arrangements would fit into the overall health sector reform efforts, and what can we learn from this experience? 3. What can we learn from successes and failures in structuring PPPs in the health sector? 4. To what extent have PPP structures taken into account the needs of the poor and their limited access to health insurance? 4 CHAPTER 1 BACKGROUND Scope and Methodology The relevant evaluation time period is FY04–15;14 the scope of the review encompasses all institutions of the Bank Group engaged in PPPs in health. The review focuses on interventions related to the provision of health services; for example, specific services, such as dialysis, or more comprehensive ones such as the operations of an entire hospital. The review did not cover Bank Group involvement in health insurance, production of pharmaceutics, or the provision or production of medical technologies. The portfolio analysis identified and categorized the characteristics, objectives, and components of the activities covered by this review and analyzed their results. Table 1.2 summarizes the Bank Group-wide portfolio of interventions with a focus or component of health PPPs. Table 1.2 Health PPP Projects Approved FY04–15 Closed/Op Health PPP Percent Institutions Mature Evaluated Projects Portfolio Evaluated by IEG Projects World Bank lending (IBRD/IDA) 25 12 9 75 IFC investments 4 1 1 25 IFC Advisory Services 49 26 16 62 Total Number of Projects 78 39 26 66 Source: IEG. The evaluation questions were answered from a desk-based portfolio review of Bank Group projects. This review relied entirely on existing evaluative evidence, primarily project evaluation data. Accordingly, IEG analyzed the results achieved at project closure for IFC Advisory Services and World Bank lending projects, and at the point of operational maturity for IFC investments. For World Bank projects, Implementation Completion and Results (ICRs) and their IEG reviews are the primary source of information on results. For IFC AS and Investment Services, this review will rely on Project Completion Reports and on one Expanded Project Supervision Report (XPSR),15 conducted at closure of the advisory mandate or at operational maturity of the investments, usually about two years after financial closure. In addition, four IFC-commissioned post-completion reports are available and were used to derive lessons from the implementation of PPP arrangements, with the understanding that these reports have not been independently validated by IEG. The portfolio review was both quantitative and qualitative. The quantitative review involved the analysis of potential trends and patterns in the occurrence of PPPs in Bank Group member states. The qualitative analysis aimed at identifying drivers of success and failure—what works and what does not work in the structuring of PPPs—and embedding them in the country health sector context. 14 Note that no health-related Public-Private Infrastructure Advisory Facility projects were identified by IEG. 15 Only one of the four IFC investments has reached operational maturity and been evaluated. 5 CHAPTER 1 BACKGROUND This approach has its limitations. Project Completion Reports and Expanded Project Supervision Reports are project-level evaluation reports with the purpose of capturing outcomes in a structured manner along certain defined dimensions and at a quite early stage, rather than at the end of PPP contracts, which often range from 5–25 years after commercial closure. Accordingly, PCRs tend to shed light mostly on the structuring process and the interaction of IFC AS teams with client governments, and to report on outcomes up to closure of the PPP. Success for AS projects is defined as a PPP project reaching commercial closure. The case is similar for ICRs, where outcomes are reported at the project closure stage, after a World Bank loan has been disbursed. Success is defined by the project development objective, often associated with the PPP reaching closure as well. However, little is known about the implementation of PPPs beyond contract closure, and even less is known about their effects on the twin goals. This was already identified as a major deficiency of Bank Group-wide monitoring and evaluation (M&E) systems in the 2013 PPP macro evaluation (see recommendation to improve monitoring of PPP performance). Moreover, evidence from outside the Bank Group on the success or failure of PPPs and their impacts on the poor is rather weak. The findings in this review have to be seen in this context, that is, as an assessment with a narrowly defined concept of success and without the ambition to pass a judgment on the success or failure of PPP’s or their effects on shared prosperity or reducing poverty. This approach does not easily lend itself to assessing performance in leveraging synergies and enabling coordination across the Bank Group. The cited project evaluation reports contain little information on Bank Group-wide coordination and collaboration; field visits would be needed to assess these two aspects properly. To circumvent—at least partially—this limitation, this review analyzed the extent to which Bank Group interventions were adequately sequenced. Adequate sequencing of interventions would allow the country to acquire the necessary skills, for example through World Bank technical assistance, or to take strategic decisions as to whether, and under what contractual arrangements, to allow the private sector to deliver health services. Subsequently, opportunities for private sector participation would be identified, for example, by World Bank–supported sector reform and strategy work. Eventually, PPPs would have to be structured, with assistance from IFC AS. Once structuring reaches commercial closure, PPPs require finance, for example from IFC Investment Services (IS). Sequencing of Bank Group interventions was assessed based on project design features, and by analyzing project components and the extent to which they supported the development of a PPP market in a country. 6 2. World Bank Group Portfolio Bank Group support covers a diverse range of activities ranging from upstream policy and strategy advice to downstream support for individual PPP transactions. More projects (58 percent) work on downstream issues than on upstream issues (42 percent), largely reflecting the relatively high numbers of IFC advisory projects. Note, however, that not all PPP-like arrangements may be captured for the World Bank because contracting arrangement and performance-based contracts may not be coded as PPPs (Table 2.1). Table 2.1. World Bank Group – Components of Assistance to Client Countries Source: IEG. Note: AS = Advisory Services; IS =Investment Services; PSP = private sector participation. Figures do not necessarily add up because these statistics rely on components within projects rather than projects. A snapshot of the Bank Group portfolio of health PPPs confirms the typical division of labor between the World Bank and IFC. Upstream support for health PPPs is mainly provided by the World Bank. All three institutions provide downstream support. While IFC AS tends of focus on structuring PPPs, that is, on downstream support, they also engage to a limited extent in upstream work: of the 49 IFC AS mandates, six provided market studies, two provided capacity building, and 42 provided advice on structuring.16 The majority of Bank Group health PPPs supported the combined health service provision (48 percent). Of this group, the large majority (60 percent) supported the provision of selective services (such as diagnostic, imaging, dialysis or radiotherapy) followed by the provision of general hospital services (30 percent). The second-largest share are health-services-only PPPs (42 percent) supported a large variety of health services including packages of essential health services (such as maternal and child health or HIV/AIDS services). The lowest share of 16Note that projects can have multiple components, providing, for example, both structuring advice and capacity building. 7 CHAPTER 2 WORLD BANK GROUP PORTFOLIO projects (10 percent) focused on facility finance, supporting only the construction and the operation of non-clinical services (Table 2.2). Table 2.2. Types of PPP Models Supported PPP model Type of health services provided Type of facility: Facility finance, accommodation only  General hospitals: 41 (10 percent)  Primary care clinics: 1 Services provided:  General hospital services: 5 Health services only (selective)  Diagnostic, imaging, dialysis or radiotherapy: 52 (42 percent)  Package of Essential Health Care Services (maternal and child health, HIV): 7 Type of facility and services:  General hospital:7 Combined health services provision  Diagnostic, imaging, dialysis or radiotherapy: 14 (48 percent)  Maternal and child health care and diagnostic and imaging: 1  Hospital-based and primary care services: 1 Source: IEG. 1 One PPP supported both the maternity wing and diagnostics of a general regional hospital. 2 One PPP also provided day-care services in addition to diagnostic, imaging, dialysis or radiotherapy. The World Bank’s upstream support focused on capacity building. About 60 percent of the World Bank’s upstream work focused on capacity-building efforts. Strategy and legal and regulatory frameworks for PPPs were the second most frequently addressed enabling factors by World Bank. The focus on capacity building has been increasingly pronounced in the past four years. Between FY11 and FY15, there were 13 instances of capacity building; by contrast, FY08-11 and FY04-07 had only five instances each. This increase represents a much-needed effort to strengthen countries’ ability to assess the appropriateness of deploying PPPs in the health sector or how to structure such engagements. The 2013 macro evaluation of PPPs had identified the lack of local skills and resources for the preparation of a PPP pipeline as a serious limitation across most Bank Group–supported countries. The Bank Group supported downstream work in increasing numbers through advising on transactions and, to a much lesser extent, through investment projects. IFC Advisory Services took the lion’s share, providing about 78 percent of all downstream support, bringing health PPP transactions to commercial and financial closure. The World Bank provided 15 percent of downstream support while IFC IS provided a mere 7 percent of it, bringing financing only to a few projects. Over time, Bank Group support exhibited significant variability. As indicated in Figure 2.1, the number of Bank Group interventions related to health PPPs ranges from two to 15 per 8 CHAPTER 2 WORLD BANK GROUP PORTFOLIO year. Clearly, the global economic crisis and the resulting credit crunch left its trace in a downturn around 2009. The subsequent increase in the Bank Group’s portfolio may be interpreted as a recovery from the crisis, but also as a sign that the 2008 World Bank Group Health Development Strategy, which called for an increased role for the private sector, was being implemented. Yet, we see an abrupt drop in 2014 and 2015. The evaluative evidence does not provide an explanation for this sudden drop. Figure 2.1. Bank Group Health PPP Portfolio over Time Source: IEG. Notes: AS = Advisory Services; IS = Investment Services. 9 3. Relevance of World Bank Group Support Bank Group support for health PPPs was relevant to countries in so far as it supported clear development priorities. PPPs addressed issues that were high on the agenda or the country development strategy as expressed by the CAS or other related strategy documents. In all of the 78 countries reviewed in detail, the Bank Group’s support for Health PPPs has been relevant to the countries’ respective developmental priorities, either directly or indirectly. In more than two-thirds of these cases (71 percent), the Bank Group’s PPP intervention directly addressed a development priority and the priority of the intervention in the country was considered highly relevant. These are cases where the intervention’s logic allowed for a direct link between the PPP and the country-specific deficiency. This is not surprising given that most developing countries regard improving the provision of health services a high priority. PPPs may be deployed to address specific health services as one of a country’s policy options. PPPs often address health needs that—given the deficiencies in a country’s health system— appear peripheral. On the assumption that PPPs deliver a better service than public provision of the same services,17 a government may resort to PPPs for the delivery of specific health services: for example, dialysis, or imaging, or laboratory technology services. As long as these assumptions hold—that is, fiscal space is created for the government or service delivery quality is increased—having the private sector provide these services, including through PPPs, is one of the government’s available policy options. Therefore, it may be unrealistic to expect that all PPPs address the most urgent health needs in a country. Two examples where IEG could assess the extent to which PPPs address health needs within countries’ specific health development agendas are explained in Box 3.1. Another way of assessing the strategic relevance of resource deployment is to see whether the Bank Group provides appropriate support to specific country needs. The Bank Group supports health PPP projects in 76 countries. Each of these countries has reached a specific level of maturity18 with regard to managing PPPs. Countries with less developed enabling environments will need fundamentally different support than would be needed by countries that already have a track record of implementing PPP projects. Nascent countries will appreciate advice on strategic issues: whether and how to use PPPs in the context of their national public investment planning and ongoing sector reform programs. The group of 17 Either because it delivers the same services at a lower cost, creating fiscal space for the government or because the services, at the same cost level, are of better quality. 18 The Economist Intelligence Unit (EIU) classifies countries with regard to PPP maturity according to a standardized procedure: “Nascent” countries are those with the least developed enabling environment; “emerging” are those where the enabling environment is under construction and less tested; and “mature” PPP countries are those that already have a quite well-established enabling environment. 10 CHAPTER 3 RELEVANCE OF WORLD BANK GROUP SUPPORT ”emerging” or “mature” countries will appreciate assistance in creating and deepening the PPP market, which IFC AS and investments can assist with. Box 3.1. Examples of Addressing Health Needs through PPPs In Burundi (P109964), one of the poorest countries in Africa, the World Bank and the government began a policy dialogue in 2000 on how to control the spread of the HIV/AIDS epidemic and reduce its impact. This dialogue resulted in the National HIV/AIDS Control Strategy (2002–06), which became the basis for the Second Multi-sectoral HIV/AIDS project of 2008. Thus, this project addressed the alarming increase in incidence rates of HIV in the country, at a time when these had stabilized in most other countries in Africa, and therefore it was highly relevant. It financed performance-based service contracts and agreements with public and private health facilities at the local level to provide clinical care for HIV/AIDS patients. The Country Partnership and Country Assistance Strategies embedded public-private partnerships (PPPs) in this program. In Turkey (IFC IS 33677), as part of the PPP developed by the Turkish Ministry of Health. This project is addressing the government’s need to renovate and enhance the quality and efficiency of healthcare services in state-owned hospitals by consolidating small state-owned hospitals under larger campuses; therefore it is of relevance to the country’s need. As a result, the IFC is assisting in the financing of the construction, operation, and maintenance of an integrated health campus in Etlik, Ankara. Etlik Health Campus will comprise 11 different buildings with a total capacity of 3,566 beds. In parallel, the World Bank is providing support through its lending program of downstream policy support in terms of capacity building of the PPP unit, and contract monitoring. There is also cross-collaboration with other international financial institutions, among them the European Bank for Reconstruction and Development, on this aspect. Source: IEG. As expected, the World Bank engages mostly nascent countries, whereas IFC supports countries in more emerging and mature environments. Forty-four percent of World Bank projects engage in nascent countries, typically helping them to create the enabling environment for PPPs. In contrast, IFC Investments and AS flow more to countries with emerging and mature frameworks: 36 percent of Investments and 40 percent of AS to emerging countries, and 40 percent to mature countries, whereas 23 percent goes to nascent countries. The difference is understandable: World Bank projects mostly support upstream activities such as capacity building, sector reform, and legal and regulatory frameworks; PPP transactions are typically supported in countries where the enabling environment is ready for their implementation (Figure 3.1). 11 CHAPTER 3 RELEVANCE OF WORLD BANK GROUP SUPPORT Figure 3.1. Distribution of Health PPP Interventions by Level of Maturity of Client Countries Source: IEG and EIU Infrascope data. Interestingly, when Bank Group support for health PPPs is compared with its support of infrastructure PPPs, a higher share of health PPP projects goes to nascent countries than is the case in the infrastructure portfolio of projects; that is, to countries with less developed PPP frameworks and a very limited record of using PPPs. Twenty-three percent of IFC AS projects are in nascent countries, which is a relatively high share, given that only 5 percent of its infrastructure PPPs are located in nascent countries. Similarly, 44 percent of World Bank health PPP support is in nascent countries, compared to 19 percent for infrastructure PPPs (Figure 3.2). 12 CHAPTER 3 RELEVANCE OF WORLD BANK GROUP SUPPORT Figure 3.2. Comparison of PPP Support to Health vs. Infrastructure by Country Maturity by EIU country classification A. Percentage Distribution of IFC AS PPP projects B. Percentage Distribution of WB PPP projects by EIU country classification Source: IEG and EIU Infrascope data. Overall, the analysis indicates that the type of health PPPs supported by the Bank Group tend to be more short-term arrangements, compared to infrastructure PPPs in the transport, energy, or water sectors. IFC AS supports governments typically with combined health service provision which includes construction, operation and provision of clinical services. While this type of PPP seems the most complex, there is significant variation within the category. On the lower risk (left side) it can involve a 10-year concession for the construction and provision of a narrowly defined health service unit (i.e. diagnostic and imaging unit). On the higher end (right side) of the risk spectrum could include a 20-year concession for the construction and operation of a hospital and clinics and include a broad range of health services provision. The review found that this latter type of combined service provision through Bank Group–supported PPPs—such as in Lesotho—are rare. IFC AS also support governments typically with health service and facility finance PPPs. These two types are on the lower-risk (left) side of the risk-sharing spectrum than combined service delivery models. Examples of such lower risk health PPPs are short-term contracts and performance-based contracts without investment of much private capital. Slightly to the right of these on the risk spectrum are the facility finance or Private Finance Initiative arrangements—these include typically the design, building, financing, and operation of a health facility (Figure 3.3A). World Bank lending provides mostly upstream support to PPP but is engaged also in supporting governments with health service PPPs through performance based contracts. (Figure 3.3B). 13 CHAPTER 3 RELEVANCE OF WORLD BANK GROUP SUPPORT Figure 3.3 Type of PPP Models Used across Country Maturity Levels (Percent of Number of Projects) A. IFC AS B. WB Lending Sources: IEG and EIU Infrascope data. 14 4. Coordination across the World Bank Group As the previous sections indicate, the World Bank’s upstream support and the IFC’s downstream support to health PPPs are potentially complementary. Table 4.1 suggests a complementarity at the aggregate level; it would be necessary to verify that such coordination and complementarity are deployed effectively at the country level. This section aims at examining the degree of complementarity of the support provided by World Bank Health, Nutrition, and Population (HNP) global practice with IFC-supported health PPPs. However, this review cannot assess actual collaboration or coordination across the Bank Group institutions because this would require field missions and would exceed the scope of the review: assessing the adequacy of sequencing of interventions. To this end IEG reviewed all World Bank lending projects19 that preceded or followed support for an actual PPP transaction. The type of support was ranked in six levels, ranging from general health sector reform to a more specific PPP type of support. To examine adequacy of sequencing, IEG identified all HNP lending operations in the 26 countries where IFC-supported health PPPs were undertaken. These lending operations were then classified into four groups: (i) projects with a component or subcomponent that explicitly supported PPPs; (ii) projects with a component or subcomponent aimed at private sector development or private sector participation in health; (iii) general health sector reform; and (iv) “other” types of HNP support.20 A total of 82 HNP operations were identified in the 26 countries during the 2003–15 period. The largest category was represented by other type of HNP support, which are projects addressing specific diseases and conditions (for example, HIV/AIDS, maternal and child health), followed by general health sector reform projects. Only seven projects included explicit components or sub-components providing support to PPPs and only one project, in Bangladesh, supported private sector development (Table 4.1).21 19 The review excludes analytical work provided by the World Bank such as analytic and advisory activities, and Economic and Sector Work. 20 The review excludes Advisory Services and Analytics support provided by the HNP and support provided by other development partners. 21 This review excludes some countries with decentralize health service delivery models such as India. 15 CHAPTER 4 COORDINATION ACROSS THE WORLD BANK GROUP Table 4.1. HNP Lending in Countries where IFC Supported Health PPPs Type of HNP support Number of projects HSR/PPP support 7 HSR/PSD support 1 General HSR 36 Other type of HNP support 38 Countries without HNP lending 2 Source: IEG. Notes: HSR = health sector reform; PSD = private sector development. The analysis of World Bank HNP support in countries with downstream IFC-supported health PPP engagements shows that adequate sequencing is rare (Figure 4.2). Detailed results are presented in Appendix 1: they show that in most of the countries, there is some type of general health sector reform (HSR) and “other” HNP support. Only two countries, South Africa and the United Arab Emirates, did not benefit from any HNP lending during the period examined. However, when it comes to preparing the country for private sector development or specific PPP support, in only a few cases do countries appear to have received timely support (see Appendix for more details on timing of support). Naturally, support for preparation is even more important for nascent and emerging countries. Of the nascent countries only Moldova and Lesotho are good examples of sequencing. Among the emerging countries specific World Bank support to health PPPs is found only in Ghana and Bangladesh. Finally, Turkey is the country were World Bank HNP has been supporting PPP capacity as part of the HSR program for several years before IFC investments took place. These results are in line with the call for a more coherent approach to realizing Bank Group synergies at the institutional level. In the 2014 health technical briefing, management recognized the need to have a more cohesive approach and pointed out each institution’s lack of familiarity with the other’s approaches to health, the lack of communication, and the need for clarity of roles and incentives. If these issues are addressed the Bank Group will be in a better position to ensure alignment and benefit from each institution’s comparative advantages. This analysis cannot take into account that other agencies may have supported one or more of these countries with adequate PSP support, or that the capacity to arrange PPPs was already present at the time of the Bank Group’s downstream engagement. This review focuses on Bank Group support. Although the Bank Group is one of the few players offering services and products along the entire value chain of deploying PPPs, as stressed in the 2013 IEG PPP 16 CHAPTER 4 COORDINATION ACROSS THE WORLD BANK GROUP evaluation,22 it is not the only multilateral agency offering support. To complete the above analysis, fieldwork would be needed to assess whether and to what extent other agencies have supported countries in developing an adequate strategy for private sector participation in health that may have benefited subsequent Bank Group interventions. Figure 4.2. Sequencing of World Bank Group Health PPP Interventions in Countries with Downstream Support Sources: IEG and EIU Infrascope data. 22IEG evaluation, World Bank Group Support to PPPs: Lessons from Experience in Client Countries FY2012. 17 5. IFC Advisory Services: Drivers of Success and Failure 23 As with any other PPP, the focus of IFC AS in health is to bring PPP transactions to commercial and financial closure. In practice, this means a two-phase process: In the first phase an options report and recommendations on transaction structure is carried out. In the second phase IFC typically helps organize a transparent, competitive bidding process that ends in a successful bid and award of concession (“contract closure”). It is desirable, though not always easy, to guarantee that the winning bidder will be able to secure financing (“financial closure”). IEG reviewed all closed projects and assessed the success of IFC AS up to the point of bringing the PPP transaction to contract closure. However, with the exception of four post-completion IFC commissioned reports,24 there is little evidence with which to measure the success of the actual PPP itself. LESSONS FROM THE STRUCTURING PROCESS Sixty three percent of all IFC AS mandates reached contract closure. Looking at the health PPP project cycle (Figure 5.1), Phase 1 was completed in 96 percent of all mandates, 83 percent of all mandates proceeded to tender with support from IFC AS, bids were received in 67 percent of all mandates, and 63 percent of all mandates proceeded to commercial closure. These data, however, are not directly comparable to data for infrastructure PPPs because a different methodology was used compared to the 2013 PPP evaluation.25 Among projects that led to contract closure, the largest success factor are project design and government commitment. Project design factors were cited in 76 percent of the successful cases, followed by government commitment, found to be a success factor in 72 percent of the projects. An important aspect of success is the willingness or the capacity of governments to undertake PPPs. Similarly, government capacity or commitment is the main reason PPPs do not pass from the options report stage to the bidding assistance stage. Three projects exemplify the importance of government decision-making and commitment for the success of PPPs. In the first example, the government decided not to proceed with the PPP models presented by the IFC AS. The Project Completion Report for the project indicates that "the problem lay in the inability of the decision maker to fully grasp and implement the 23 The number of PPP projects having evaluative evidence for IFC Investment Services and the World Bank is very limited, so patterns and lessons to be learned can only be distilled for IFC Advisory Services (AS) projects, where a larger sample of evaluated PPP projects is available for study. 24 IFC began to conduct post-completion evaluations in 2012. 25 For Health PPPs, to be in line with IFC AS’ own calculation methods, IEG calculations for this study now include active projects that reached commercial closure but still showed as active in the system, and terminated projects; both of these categories had previously been excluded. 18 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE core PPP Principle, which is the financial interaction between the private operation and the state budget….” In the second example, the government stopped the project even before reading the PSP plan because there was insufficient support in the country for the proposed reform. In the last example, the bidding process was cancelled not by the ministry of health, but by the ministry of finance. A meeting set up with the minister of health to try to resolve the problem was cancelled by the government. IFC tried, but was never able to confirm interest in pursuing the project from the new ministries of health and finance, and after having the project on hold for five semesters, IFC decided to terminate it, following a change in government. Figure 5.1. IFC Advisory Success along the Health PPP Delivery Chain Sources: IEG, Project Completion Reports. Note: n = 24. IEG identified 49 PPP- targeted transactions by IFC Advisory Services, approved during FY 04–15, of which 41 are transaction advice with the objective of commercial/contract closure (or commercial closure in another term). Of these, 24 have reached contract closure or have been terminated. The remainder are active, or are studies that do not aim at contract closure of a specific transaction. A 63 percent share of PPP advisory mandates reaching commercial closure should not be interpreted as low. First, success depends, as shown, to an overwhelming extent on external factors, mainly government commitment or the availability of a champion. Second, a relatively lower share of PPP mandates reaching commercial closure compared to those initiated can also be interpreted as IFC AS “pushing the envelope;” that is, IFC is trying to bring PPP structures to countries with a relatively low PPP track record—a potential indication of the pioneering nature of IFC’s support. However, this needs to be viewed in context, because governments in nascent countries are likely to lack the capacity and skills to structure and manage PPP contracts in a way that safeguards the public interest. Because evidence is slim on the effects of PPPs beyond closure, particularly on their contribution to 19 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE the Twin Goals and serving the poor, the verdict on their role in bringing services to poor countries is still out. Bank Group advice has not been strategic enough with respect to laying out all options for the provision of health services in a given country. IEG reviewed the extent to which IFC AS, in its phase 1 analysis, presented public procurement as an option; of 12 Strategic Options Reports reviewed by IEG, public procurement was presented in one case. The demand-driven nature of IFC AS may explain this finding: Governments contract with IFC once they have decided to proceed with a PPP. In light of the above analysis indicating that the Bank Group is not adequately sequencing its interventions and that support for PSD in client countries is not provided in a timely manner, the question arises: On what basis have these governments decided to go for the PPP model? The extent to which it should be IFC’s responsibility to include the public service option as part of its assessment is open to debate; nonetheless, it should become Bank Group practice to ensure that the public option is at least considered and systematically assessed (or that it has been assessed with support of other agencies). The efficiency and desirability from a social perspective of the PPP cannot be established without a comparison with the alternatives, the main one being the public option. This needs to be done ex-ante, during and ex-post the PPP transaction. The Bank Group may consider how to engage with governments in a more strategic manner and provide advice on whether and how to best use PPPs in a health context—a recommendation already issued by the 2013 IEG evaluation. PPPs may have fiscal implications for the public—an important aspect to consider when they are being structured. Fiscal implications may result from investments of the public sector in the preparation of the PPP, that is, in providing the infrastructure in the form of buildings and other needed assets; they may also originate from payments that are due once the contract is concluded. Such payments may take the form of availability payments; for example, for the provision of a minimum standard of service, or payments that are a function of the units of service provided. For substantial investments, assessing fiscal implications is good practice; however, the 2013 PPP evaluation found that fiscal implications, in particular contingent liabilities, are rarely assessed. This may jeopardize the fiscal sustainability of client countries. Focusing on those PPP arrangements with substantial fiscal implications, this review found little evidence that fiscal implications are assessed in a systematic fashion (Figure 5.2). Of the 18 closed PPPs, 5 had potentially no or very limited fiscal implications because they were contracts for the provision of specific health service (see Figure 5.2). For the sake of this review, it was assumed that these pose limited risk to client countries’ fiscal space; hence, the analysis focused on the remaining 13 PPPs that were facility finance or combined service delivery arrangements. Because of the complexity and magnitude of these PPPs, it appears advisable to assess fiscal implications. 20 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE Figure 5.2. PPPs and their Fiscal Implications (Projects with IFC AS advice on structuring PPPs) Source: IEG. Of the 13 PPPs with potentially significant fiscal implications, about eight there was evidence enough to make a detailed assessment of these implications (Figure 5.3). The vast majority of these PPPs relied on the government for payment of services: Fully eight of 13 PPPs relied entirely on the government for service payments; six on governments and patients’ payments; two on government and insurance schemes; and one on government, insurance, and patients (Figure 5.3). In summary, it can be concluded that almost of all these 13 PPPs substantially depended on the government. IEG could obtain evidence of structured and detailed assessment of fiscal implications for about 62 percent of the closed projects (eight of these projects). This corroborates the findings of the 2013 PPP evaluation. Following the 2013 PPP evaluation, the Bank Group, together with the IMF, developed a PPP Fiscal Risk Assessment Model (PFRAM).26 This tool, however, is yet to be implemented. Hence, only about 60 percent of the relevant PPPs had a structured assessment. It will be 26PFRAM is available at https://www.imf.org/external/np/fad/publicinvestment/pdf/PFRAM.pdf 21 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE essential to apply this tool systematically to all those PPP projects that have the potential to threaten the fiscal space of a client country. Figure 5.3. Fiscal Assessment in PPPs with Expected Substantial Fiscal Implications Source: IEG. PPPS AND THE POOR The underlying rationale for PPP interventions is that PPPs can help improve service delivery and the provision of basic infrastructure, including for the poor. To understand the extent to which PPP structures have taken into account the needs of the poor and their—usually limited— access to health insurance, IEG reviewed all projects for which a focus on the poor was stated in the project approval documents. For those projects that did not explicitly mention access for the poor in approval documents, the geographic location has been analyzed. From an understanding of how many of the supported PPPs are located in areas where poor people live it could safely be assumed that these poor would also be the beneficiaries. A third of IFC AS projects are explicit about their poverty focus in project documents; when the geographic location27 of the supported health PPPs is also considered, it could be concluded that up to two-thirds of supported PPPs have a pro-poor dimension. Results show that in 36 percent of IFC AS projects (16 out of 44) the respective PPP arrangements are explicitly designed to take the poor into account. However, when project locations are 27 A project located in a low-income state was used as a proxy for poverty focus. 22 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE analyzed, the percentage increases to 68 percent. This means that the vast majority of Health PPPs supported by IFC AS had a poverty focus and followed that logic. IFC Investments had only four projects during the same time period; though these did not specifically mention targeting the poor, they provided health services to the general public, including the poor, under a universal health care regime. The vast majority of World Bank–supported PPP interventions have an explicit emphasis on the poor or underserved populations in their design. Between FY04 and FY15, 84 percent (21 out of 25) of World Bank projects specifically targeted the poor. When the geographic dimension is added, the percentage of pro-poor World Bank PPPs increases to 92 percent. For example, in the Democratic Republic of Congo, the World Bank targeted districts that are among the poorest and most isolated in the country, financing performance-based contracts for the improvement of health services. However, although the design of the majority of Bank Group interventions has a pro-poor focus, M&E indicators are lacking to track results and to assess whether the poor were able to access the health services. IEG reviewed the indicators established at the design stage. For IFC advisory services, only four projects (out of 16 IFC AS interventions that have a pro-poor focus) have specifically pro-poor indicators that can be tracked at project completion. Examples of pro-poor indicators are related to access to service for the poor or underserved segments (that is, the number of people below the poverty line receiving the service, the percentage of people below the poverty line out of all admissions receiving the service). For the World Bank, while most of the nine closed projects had indicators related to the poor or underserved population at the design stage, the majority of indicators lacked adequate baselines or targets or were inadequate to be assessed. LESSONS FROM OPERATING PPPS From the four Project Completion Reports available on IFC AS–supported PPPs, a few lessons can be drawn as to what contributes to operating PPPs in a successful manner after commercial closure. The following paragraphs summarize those lessons, though it should be noted that these are drawn from IFC-commission evaluation reports that have not been independently validated by IEG. Establishing formal mechanisms that allow regular interactions between government and the concessionaire was key in at least one project reviewed by IEG: The concession agreement for the project required the government and the concessionaire to form a liaison committee comprising representatives from the government and the concessionaire. This committee could look at the issues relating to the day-to-day operation of the contract. Our findings suggest that this committee was very helpful in ensuring the smooth functioning of the centers at the initial stages. Robust M&E, with adequate baselines and key performance indicators are essential: In the case of a radiology project, performance specifications in the concession agreement did not 23 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE define the key performance indicators based on specific, measurable, and achievable principles. The long list of performance specifications mentioned in the concession agreement is merely a reporting requirement and does not have benchmarks to measure the performance of activities. In a dialysis project, though there were some indications of improved health outcomes in privately managed clinics, the data were insufficient to determine whether the improvements in privately managed clinics were translating into improved mortality rates for patients. Payment security reduces financial risk and enhances private sector confidence: many health projects require payment from governments to private operators. In some countries, PPP laws require that concessionaires be provided with a payment guarantee. Hence, project structures can require the government to route a part of the expenditure budget, equivalent to the maximum monthly payment, to an escrow agent. The payment due to the concessionaire (which in the example reviewed by IEG was calculated based on the Independent Verifier’s Report) was automatically transferred to the concessionaire. In another project, an escrow account or letter of credit provided payment security and ensured timely availability of funds from the government. According to the terms of the concession agreement for that project, the government was required to provide a revolving letter of credit to the concessionaire; however, after the agreement was signed, the government did not provide the letter of credit as stipulated. This exposed the concessionaire to payment delays from the government. Contractual flexibility is a good tool for midway corrections: The concession agreement for a project reviewed by IEG had the flexibility to allow reviews and revisions at regular intervals during the concession period, along with provision of extraordinary adjustments on performance targets as and when needed. Post-completion reports also indicate some positive effects of PPPs, for example on access and quality, but evidence is limited. The following paragraphs summarize the effects of PPPs along the dimensions of access, quality, efficiency, financial soundness, and access for the poor. Along several of these dimensions, data indicate progress, while in others progress has been uneven. Data were extracted from four PPPs which were supported by IFC AS until financial closure. Results have not been independently verified by IEG. The projects indicate good results with regard to access and quality of services. Three out of the four projects show evidence that access has increased compared to the baselines. In one case, the PPP project delivered significantly more services and higher quality in the first year of operation than at baseline. The number of admissions increased 51 percent, outpatient visits more than doubled, and the hospital and filter clinics assisted 45 percent more deliveries than the baseline. The second project provided services for 170,000 compared to a target of 74,000. The third project caters to 11 percent of the total number of hospitalizations versus a baseline of 8.5 percent. In addition, prior to the establishment of the hospital, the people in the area used to travel 22 km for treatment of more complex health complaints. Project quality and performance targets were generally achieved. 24 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE Reports show positive effects on indicators measuring access for the poor, but more detailed assessments are needed. All four projects targeted poor areas. In one case, in a country with the largest population of extremely poor, hospital services were free for the population. In the second example, there has been a reduction in the out-of-pocket burden for public patients, and services are provided free of cost to the poor. In the third example, according to the reports, more detailed assessments are needed to determine whether the poor people in the areas of influence are choosing not to seek care at the hospital because of inability to pay costs including any fees, transport, or other expenses associated with seeking care, including waiting time. The indicators for efficiency were mixed. Some projects exceeded benchmarks and targets, while others performed less well. Indicators were also diverse, making direct comparisons more difficult. In one case, the report points to the difficulty of demonstrating direct quantitative evidence of improved clinical efficiency as a result of providing advanced radiology services in the teaching hospitals; therefore, the measure of impact on clinical efficiency was based on qualitative information only, collected from the doctors at the hospital. In a second project, measures of efficiency and performance were mostly achieved. Most of these results show the inadequacy of the current M&E system for health PPPs—they also pose a possible reputational risk for the Bank Group. That only four studies are available on the effects of PPP beyond closure, coupled with the emerging findings in these reports that in most cases more research would be needed, indicates that the M&E system used to monitor the effectiveness of health PPPs is insufficient. The 2013 IEG PPP evaluation also identified major shortcomings in the way the Bank Group assesses the long-term effectiveness of PPPs. The corresponding recommendation, issued in 2013, is currently under implementation. Frequently, PPPs encounter issues only after a few years of operation, requiring renegotiations or adjustments to the underlying contracts. The literature indicates a high incidence rate of renegotiations of PPP contracts. According to a 2014 study by the Organization for Economic-Cooperation and Development (OECD),28 68 percent of infrastructure29 concessions in Latin America were renegotiated; typically, such renegotiations occurred within two years of the contract award. The report adds that these renegotiations can undermine the benefits of the initial competitive bidding process, because renegotiations are often bilateral, between the winning operator and the government, with limited or no transparency. 28 Guasch, J. L. et al. (2014), “The Renegotiation of PPP Contracts: An Overview of its Recent Evolution in Latin America”, International Transport Forum Discussion Papers, 2014/18, OECD Publishing, Paris. http://dx.doi.org/10.1787/5jrw2xxlks8v-en. 29 The most common sectors continue to be transport, water, and sanitation; yet social sector PPPs such as health are picking up. 25 CHAPTER 5 IFC ADVISORY SERVICES: DRIVERS OF SUCCESS AND FAILURE The main reasons for renegotiations are weak contract management and weak government institutions. The 2014 OECD report30 indicates that while post-contract award management of PPP projects is indeed critical for securing the expected benefits, it has generally not been an important priority for governments. This is indeed surprising given that most PPP contracts have durations of 5–30 years with considerable contract complexity. Moreover, monitoring of service delivery poses a big challenge for governments. Despite the need, the Bank Group rarely provides “aftercare” for contract management. Providing aftercare would also give the Bank Group an opportunity know ‘first hand’ from post-financial closure results. Though IEG is aware that IFC AS and the World Bank offer aftercare type of support, this study notes that uptake has been very limited so far. 30Guasch, J. L. et al. (2014), “The Renegotiation of PPP Contracts: An Overview of its Recent Evolution in Latin America”, International Transport Forum Discussion Papers, 2014/18, OECD Publishing, Paris. http://dx.doi.org/10.1787/5jrw2xxlks8v-en. 26 6. Lessons Learned As the development community and the World Bank Group increasingly partner with the private sector to enhance the delivery of health services, this review stresses five lessons: (1) When advising governments on the various models for providing health services, both the IFC and the World Bank should act as one Bank Group and discuss the whole range of options, from the public and mixed options to the other possible types of PPPs, in the context of the country’s state of reform, overall maturity, and track record in using PPPs. (2) To be better positioned to deliver such strategic advice, the Bank Group should better integrate its sector reform and policy work with its structuring and financing of PPPs. (3) Access for the poor and affordability need not only be systematically considered at the design stage, but also tracked to ensure that the poor actually benefit from PPPs. (4) The recently developed Bank Group–IMF PFRAM tool for assessing fiscal implications should be systematically applied to structuring PPPs that have substantial fiscal implications. (5) The new practice of preparing post-completion reports after the PPPs have gone into operations should be mainstreamed so that post-completion reports are conducted after sufficient time, given the long contractual life of PPPs. 27 Appendix. Analysis of World Bank HNP Support in Countries with Downstream IFC-Supported Health PPPs Table A.1. World Bank HNP support in nascent countries Country Maturity 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 HSR - P082814 HSR - P144688 Albania Nascent IFC-AS HNP-P096482 HNP-P143652 HNP-P073118 HNP-P096056 Benin Nascent HSR-P074841 IFC-AS HNP-P071433 HNP- P119917 Burkina Faso Nascent HNP-P093987 IFC-AS HNP- P076715 Grenada Nascent IFC-AS HSR-P051372 HSR-P126278 Kyrgyz Nascent HSR-P084977 Republic IFC-AS HNP-P107375 HNP- P087843 HNP-P114859 Lesotho Nascent HSR/PPP-P076658 PPP-P104403 IFC-AS IFC-AS FYR HSR-P086670 Nascent Macedonia IFC-AS HNP-P099841 HNP-P074122 Moldova Nascent HSR-P144892 PPP-P095250 P125719 IFC-AS HSR-P133187 HSR-P051370 HSR-P113349 Uzbekistan Nascent PPP-P125004 IFC-AS 28 Table A.2. World Bank HNP support in emerging countries Country Maturity 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 HNP-P096056 Bosnia and HSR-P088663 P120285 Emerging Herzegovina HSR-P071004 IFC-AS HNP-P080228 Egypt Emerging HSR-P045175 IFC-AS HNP- P088797 HNP-P071617 HNP-P105092 HNP-P145792 HSR-P101852 Ghana Emerging HSR-P073649 PPP-P125595 IFC-AS HNP-P082242 Honduras Emerging HSR-P053575 IFC-AS HNP-P102119 HNP-P080295 HNP- P146583 HNP-P070291 HNP- P120798 P130865 Nigeria Emerging HSR-P104405 HSR-P070290 IFC-AS IFC-AS HSR-P078971 HSR-P145174 Romania Emerging HSR-P008797 IFC-AS United Arab Emerging IFC-AS Emirates HNP-P102305 HSR-P074841 Bangladesh Emerging HSR/PSD- P118708 IFC-AS 2 Table A.3. World Bank HNP support in developed countries Country Maturity 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 HNP-P095171 Brazil - Bahia Developed HSR-P054119 IFC-AS IFC-AS IFC-AS HSR- P051273 HSR-P086669 HSR-P144871 Croatia Developed IFC-AS HNP- P116226 Mexico Developed HSR-P066321 IFC-AS HSR- P062932 HSR-P095563 Peru Developed IFC-AS South Africa Developed IFC-LEN HNP-P096262 HSR-P074053 Turkey Developed HSR/PPP-P102172 HSR/PPP-P152799 3 IFC-LEN 29