41702 noTE no. 27 ­ SEpT. 2007GRIDLINES Sharing knowledge, experiences, and innovations in public-private partnerships in infrastructure Designing and using public-private partnership units in infrastructure Lessons from case studies around the world Apurva Sanghi, Alex Sundakov, and Denzel Hankinson P ublic-private partnership (PPP) units for And where there are problems that a PPP unit facilitating and managing infrastructure cannot address, governments should consider investments have existed for years in many other institutional solutions, or resolve the issues developed countries. Driven in part by growing before creating a PPP unit. infrastructure investment, these units have also recently begun to proliferate in the developing A qualitative assessment of eight PPP units around world. While governments often seem eager to the world--in Bangladesh, Jamaica, the Republic create such units, not everyone in the global PPP of Korea, the Philippines, Portugal, South Africa, market is convinced of their value. An assess- the United Kingdom, and the Australian state of ment of eight PPP units around the world exam- Victoria--provides lessons on effective design and ines whether these institutions have contributed use of these institutions (PPIAF and World Bank to successful public-private partnerships--and if forthcoming). so, under what conditions. There has been much enthusiasm recently among Why governments create PPP units governments and donors for establishing units to implement or advise on public-private partner- Understanding the role of PPP units requires ships--or PPP units. Albania, the Arab Republic of an appreciation of the role of PPPs in achieving Egypt, Malawi, Mozambique (Maputo), Nigeria, governments' policy objectives. In many of the Tanzania, and Turkey have all recently considered case studies the government initially used PPPs establishing PPP units. to attract private finance as a way to overcome fiscal constraints. But governments that have The growing popularity of these institutional a long history with PPPs have recognized their structures is reminiscent of a similar trend in the usefulness in achieving more specific objectives: 1990s--establishing independent regulatory agen- net present value of money as measured against cies for infrastructure services such as water and services the government typically provides on its electricity. But where developing countries simply own, and optimal risk transfer to private partners transplanted successful regulatory models from (rather than maximum risk transfer to the private the United Kingdom or the United States, results sector). (See box 1 for some useful definitions.) were mixed. Good regulation requires painstaking institutional design and a clear understanding of a But achieving such objectives is no simple task. country's needs, capacity, culture, and administra- Managing a successful PPP program requires a tive traditions. Similarly, there is no "one size fits range of specialized functions, and not all govern- all" approach to designing or using PPP units. ments will have those functions or the ability Before designing a PPP unit, governments first need a clear understanding of the problems they PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY Apurva Sanghi is a senior economist in the World Bank's face in implementing their PPP program. These East Asia and Pacific Region, Operations and Policy Unit. problems should drive the design of the PPP unit. Alex Sundakov is executive director of Castalia. Denzel Hankinson is a manager at Castalia. Helping to eliminate poverty and achieve sustainable development through public-private partnerships in infrastructure PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY figure 1 functions PPP units can provide to address government failures in PPP programs Government functions Government failures Functions to address failure Set PPP policy and strategy Poor procurement Quality control incentives Project origination and identification Policy formulation Lack of coordination and coordination Analysis of individual projects Technical assistance Lack of skill Standardization Transaction management High transaction costs and dissemination Promotion Contract management, monitoring Lack of information and marketing and enforcement Source: PPIAF and World Bank forthcoming. to perform them effectively (figure 1). Where tant that the lines of responsibility be formalized governments have weaknesses in the functions and clear to public and private partners alike. If no PPP units needed to effectively manage a PPP program--or government agency is well suited to correcting the "government failures"--they often create a PPP government failures in a country, responsibility need to be unit to help in correcting those failures. The need for correcting them should fall to the PPP unit. designed to address specific government failures is one to address reason PPP units need "custom" designs. How have PPP units fared? specific In principle, each of these functions could be government performed by a line agency, a coordinating agency The case studies show that in countries with (such as a cabinet office), or external consultants. successful PPP programs, the PPP units that failures There are many institutional solutions to integrat- contributed to that success performed more of ing these functions that don't involve creating a the functions necessary to correct government PPP unit. But if a specialized organization is to be failures. Table 1 summarizes the results of this created to address the government failures, it must comparative analysis. As the table shows, there is be able to perform these functions effectively. a high positive correlation between how success- That means it needs to be given the necessary ful a country's PPP program was and how many executive authority rather than simply act as an of the functions necessary to correct government advisory body. failure a PPP unit performed. Not all functions related to PPP procurement need The units that were least effective--those in reside in a single PPP unit. The United Kingdom Bangladesh, Jamaica, and the Philippines--are in now has a separate PPP task force in Treasury as countries where indicators of government effec- well as its technical assistance and project devel- tiveness are weak relative to other countries in opment body, Partnerships UK. In addition, some the sample (PPIAF and World Bank forthcoming). line ministries have their own internal project That should be no surprise: in countries where development teams. When responsibilities are most government institutions perform poorly, new divided among agencies in this way, it is impor- institutions are also likely to perform poorly. This Private participation in infrastructure in Europe and Central Asia PPP program in its jurisdiction. To summarize box 1 from the case studies: Defining some useful terms · In Bangladesh much PPP activity has happened A public-private partnership (PPP) is an agree- without the involvement of the Infrastructure ment between a government and a private firm Investment Facilitation Center (IIFC), suggest- under which the firm delivers an asset, a service, ing that IIFC has had little impact on the or both in return for payments contingent to quantity or quality of PPPs. Less effective some extent on the long-term quality or other · In Jamaica the National Investment Bank governments characteristics of outputs delivered. Agree- (NIBJ) was the principal driver of the privatiza- ments may range from service or management tion program and therefore had much influence tend to have contracts to concession agreements and priva- over the PPP program--though it ultimately less effective tization and cover widely varying activities, not answered to the Cabinet, a political body. PPP units just those in infrastructure sectors. · In the Philippines, much as in Bangladesh, significant PPPs have happened without the A PPP unit is any organization designed to BOT Center's involvement in recent years. The power sector PPPs in which the BOT Center · Promote or improve PPPs by trying to attract was directly involved in the 1990s varied in more of them or by ensuring that PPPs meet quality and have left the country with signifi- such quality criteria as affordability, value for cant liabilities. money, and appropriate transfer of risk. · In Portugal Parpública is the principal driver · Have a lasting mandate to manage multiple of PPP policy and has close links to the Trea- PPP transactions, often in multiple sectors. sury, which has driven fiscal reform since 1999. This distinguishes PPP units from PPP teams Parpública has had much to do with improving working in a single ministry or committees the affordability and value for money of PPPs while allowing the deal flow to remain relatively created to work on specific transactions. high. A successful PPP unit is one that contributes to · In South Africa the Treasury's PPP unit plays a a successful PPP program. central part in developing PPPs. Though some- times criticized as too restrictive, the unit was A successful PPP program is one that fosters created with an inherently restrictive aim of successive PPP transactions that ensuring that PPPs happen--but not as a way of avoiding budgetary constraints. Its regulations, · Provide the services the government needs. its manual, and many of its completed transac- · offer value for money as measured against tions are cited outside South Africa as examples public provision of services (with value for of good practice. money measured by the net present value of · In the United Kingdom and the Australian state lifetime costs, including the cost of bearing of Victoria, two of the world's largest markets risk). for PPPs, the PPP units--Partnerships UK and Partnerships Victoria--are central to the PPP · Comply with general standards of good gover- programs. nance and with such policies as avoiding · In Korea the Public and Private Investment corruption, being fiscally prudent, and comply- Management Center (PIMAC) plays an essen- ing with relevant legal and regulatory regimes. tial part in evaluating feasibility studies and bids. Private participation in infrastructure has picked up considerably since the government created PIMAC's predecessor, the Private Infra- conclusion may seem obvious, but it has some structure Investment Center of Korea (PICKO), significance for the decision about how to set up in 1999. a PPP unit. It suggests that a PPP unit may not be able to provide an "island of excellence" within a government that is generally poorly performing. Lessons Correlation does not, of course, mean causation. The case studies point to some lessons on the But the case studies provide at least anecdotal appropriate design and use of PPP units in infra- support for the effect each PPP unit had on the structure--and some reasons for the positive correlation between successful PPP programs and · A PPP unit's location in the government is among the use of PPP units. the most important design features, because of the importance of interagency coordination and · Less effective governments tend to have less effec- political support for a PPP unit's objectives. tive PPP units. Lack of political commitment In a parliamentary system of government a to advancing a PPP program, or lack of trans- PPP unit is most likely to be effective if located parency and coordination within government in a strong ministry of finance or treasury. In agencies, will reduce the chances of success for a nonparliamentary systems, such as the presi- PPP unit. Even with a good design, a PPP unit is dential system of the Philippines and many unlikely to be effective in such an environment. Latin American countries, the best location for The least effective PPP units are in countries a PPP unit is less clear. In a country with a whose governments as a whole are relatively less strong planning or policy coordination agency, effective. that agency might make a natural home for a · Without high-level political support for the PPP PPP unit. program, a PPP unit will most likely fail. · Relatively successful PPP units directly target specific Policy makers and their advisers need to consider government failures. A clear focus on responding these fundamental lessons before turning to any to particular government failures is essential in of the more detailed considerations about a PPP ensuring the success of the institutional solution unit, such as its structure or staffing. Thinking selected. about PPP units should begin with questions · The authority of a PPP unit must match what it is about what government failures need to be expected to achieve. If a PPP unit is expected to addressed and then whether (and, if so, how) the provide quality control or assurance, it needs the unit can be given sufficient influence to address authority to stop or alter a PPP that it perceives those failures. to be poorly designed. But this executive power must be coupled with a mandate to promote Reference good PPPs--or the unit may simply wield a veto PPIAF and World Bank. Forthcoming. PPP Units for Infrastructure: without adding value. Lessons for Their Design and Use. Washington, DC: PPIAF and World Bank. GRIDLINES Gridlines share emerging knowledge on public-private partnership and give an overview of a wide selection of projects from various regions of the world. Past notes can be found at www.ppiaf.org/gridlines. Gridlines are a publication of PPIAF (Public-Private Infrastructure PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY Advisory Facility), a multidonor technical assistance facility. Through technical assistance and knowledge dissemination PPIAF supports the efforts of policy makers, nongovernmental organizations, research institutions, and others in designing and implementing strategies to tap the full potential of private involvement in c/o The World Bank, 1818 H St., N.W., Washington, DC 20433, USA infrastructure. The views are those of the authors and do PHone (+1) 202 458 5588 fAx (+1) 202 522 7466 not necessarily reflect the views or the policy of PPIAF, PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY generAL eMAiL ppiaf@ppiaf.org Web www.ppiaf.org the World Bank, or any other affiliated organization. TAbLe 1 Summary of success or failure of PPP units How many functions Did necessary for solving How much PPP unit government failure? Jurisdiction success did PPP What were meet its Did PPP unit perform and unita program achieve? PPP unit's objectives? objectives? effectively? Bangladesh Little success · Advise line ministries and No apparent None to few IIFC (1999) government agencies in effect on · Technical assistance identifying, evaluating, private parti- · Policy formulation, awarding, negotiating, cipation in but no implementa- and implementing projects infrastructure tion authority · Promote private participa- tion in infrastructure and serve as clearinghouse of expertise on PPPs Jamaica Little success · Secure greater efficiency No, especially None to few NIBJ (1988) · Reduce fiscal drain not the reduc- Managed some · Optimize government's tion in fiscal transactions, but real management resources drain power never effectively · Secure enhanced access to delegated foreign markets, technology, and capital · Broaden ownership Portugal Much success · Help structure higher-quality Yes Some Parpública PPPs · Policy formulation (mid-1990s) · Technical assistance · Quality control South Africa Much success, · Filter out fiscally irrespon- Yes, but scant Some Treasury PPP despite low deal sible PPPs while creating a impact on · Technical assistance unit (2000) flow structure for PPPs that would infrastructure · Quality control reassure private investors deals · Policy formulation despite its being a fine filter Republic of Much success · Provide technical assistance Yes Most to all Korea PIMAC to government agencies and · Technical assistance (2005) private sector · Quality control · Promote infrastructure pro- · Policy formulation jects and educate private · Promotion and sector about PPPs marketing · Review unsolicited propo- sals, feasibility studies, and bidding documents · Conduct value-for-money tests, evaluations, and nego- tiations; formulate PPP policy Philippines Some success, · Provide technical assistance Yes, but not for None to few BOT Center though IPPs of the · Promote and market PPPs all PPPs Assigned many func- (1993) 1990s left significant · Monitor PPPs tions but effective only contingent liabilities in technical assistance United Much success · Improve quality of PPPs Yes Most to all Kingdom · Shift focus away from finan- · Technical assistance Partnerships cing infrastructure to value · Quality control UK (1996) for money and risk allocation · Policy formulation and Treasury · Promotion and task force marketing Victoria, Much success · Improve quality of PPPs in Yes Most to all Australia infrastructure · Technical assistance Partnerships · Ensure that PPPs provide for · Quality control Victoria (1999) optimal risk transfer, maxi- · Policy formulation mize efficiency, and minimize · Promotion and lifetime costs marketing Source: PPIAF and World Bank forthcoming. www.PPIAF.org. a. The year in parentheses is the year the unit was established.