Are public–private partnerships (PPPs) the answer to Africa's infrastructure needs?

Introduction

There is a large and growing infrastructure gap in Africa which has, inevitably, led to the exploration of alternative ways of financing infrastructural development. Public–private partnerships (PPPs) have been proposed as a possible major solution. Multilateral bodies within Africa, such as the African Union (AU), the UN Economic Commission for Africa (UNECA) and the African Development Bank (AfDB) have all endorsed their use, prompted by financial and technical support from the World Bank, the OCED (2012), and by endorsement, albeit a cautious one, by the IMF (IMF, 2004). This paper explores the nature of PPPs, the extent of their use and their location by sector in Africa. It also examines the arguments advanced for the promotion of PPPs and looks critically at them. It concludes that great caution should be exercised in the use of PPPs.

The infrastructure gap in Africa

Annual infrastructure needs in sub-Saharan Africa (SSA) have been estimated to be US$93 billion (Foster and Briceño-Garmendia 2010, 6–9, 65–86). Table 1 below shows that US$33.0 billion or about one-third of annual infrastructure spending needs is on operations and maintenance while US$60 billion is required for capital expenditure. The biggest need is in power, followed by water and sanitation and then transport.

Table 2 gives an indication of the infrastructure deficiency in low-income and middle-income African countries compared with countries at similar income levels elsewhere. It can be seen that Africa lags in all areas of need except unpaved roads. The deficiency is particularly glaring with regard to paved roads, landline phones, the Internet and power availability.

Of the annual infrastructure needs of US\)93 billion, only US$45 billion is available through existing sources (Table 3) leaving an ‘infrastructure gap’ of US$48 billion equivalent to 4.2% of GDP. Furthermore, it is widely accepted that the infrastructure needs of African countries are growing.

The largest source of funding for infrastructure is the public sector, which provides almost US\)30 billion, or two-thirds of all funding. Foreign aid provides about US$6 billion or 13% (with China significantly increasing its contribution in recent years) and the private sector US$9.4 billion or 21%. There is widespread scepticism about the ability of the public sector to meet the infrastructure deficit, even with foreign aid. This, together with the current preoccupation of many governments and international organisations with neoliberal approaches to the provision of public services, has led to the promotion of PPPs as a possible solution to financing the infrastructure shortfall.

Public–private partnerships (PPPs): what are they?

In conventional procurement private sector contractors almost always build the project while the public sector owns, arranges design, financing, operations and maintenance. In PPPs, there is usually a long-term contract between a private party and a government agency for some combination of construction, ownership, design, financing, operation and/or maintenance of public infrastructure. In other words, the private sector becomes more involved in public projects, taking over functions traditionally retained by the public sector. For this reason, PPPs have been described by their detractors as privatisation by stealth (CUPE 1998). Proponents, however, see them in a more benevolent light as being cooperative arrangements ‘between the public and private sectors, built on the expertise of each partner, to develop or improve facilities, infrastructure and/or operating services on behalf of the public, through the appropriate and fair allocation of resources, risk, rewards and responsibilities’ (City of Winnipeg 1999, 6)

Figure 1 indicates the range of arrangements that qualify as PPPs, from the least private sector involvement at the top to almost complete privatisation at the bottom. In the least invasive forms of PPP, the operations and maintenance functions can be handed over to the private sector while all other aspects of infrastructure projects remain in public hands. Greater private involvement might then involve the private sector providing the finance directly, through loans and private equity. At the top end, before complete privatisation, public infrastructure can be run as a long-term concession in which the private sector designs, builds, finances, owns, operates maintains and possibly charges the public for the use of infrastructure. In the last 20 years, many developing countries have also increasingly involved the private sector in rehabilitating public infrastructure.Figure 1.

Figure 1.

The PPP spectrum.

In a survey of PPPs in developing countries between 1990 and 2003, Hammami, Ruhashyankiko, and Yehoue1 (2006, 12) found that three types of PPP accounted for 70% of all projects: build-own-operate PPPs (BOO) accounted for 38.9%, build-own-transfer (BOT) for 17.9% and build-rehabilitate-operate-transfer (BROT) for 13.2%.

The attraction for the private sector is that the public sector pays it an annual lease for using infrastructure, which is then used to repay borrowing, cover costs and make profits. PPPs, in effect, enable greater commodification of public services by extending the involvement of private capital in the sphere of public service. Contracts are often very long-term, from 25 to 40 years, and direct ownership of infrastructure is more common in Africa than in developed economies. The private companies involved earn large returns on their equity investments. Governments replace direct servicing of loans with the payment of annual leases.

Extent of private involvement in African infrastructure

Accurate figures of PPP activity in Africa are impossible to come by. Some indication is given by the World Bank's Private Participation in Infrastructure Database (World Bank 2013a). This shows that between 1990 and 2011, there was US$121 billion of private funding of infrastructure in SSA, just over 10% of the total such involvement in developing countries as a whole. US$95 billion or 78% was in the telecom sector, US$14 billion in transport and US$12.5 billion in energy. Only US$0.3 was invested in water and sanitation, as this sector does not appear to be attractive to private investment in Africa.¹ Not all of this investment, however, was in PPPs, for US$24 billion was the outcome of divestment where the public facility was simply privatised. The weaknesses of the data are, however, many. To begin with they cover only large projects, but more importantly they exclude projects in health, education, prisons and simple government buildings. In South Africa, for instance, investments in PPPs in such sectors amounted to at least Rand 1.4 billion in 27 projects by 2011 (South Africa 2013). There were only four PPPs in sectors covered by the World Bank data though admittedly, at Rand 3.8 billion, they were much larger projects. The World Bank database is, therefore, likely to underestimate the number of PPPs in Africa and the scope of investment in them.

Nonetheless, still using that data, it appears that South Africa and Nigeria account for 87 projects out of an estimated 436 in SSA, or about 20%, and at US$63 billion, for over a half of total investments. There are 11 other SSA countries each with US$2 billion or more in PPPs, accounting for 184 projects. In North Africa, there were 75 identified projects costing US$65 billion.

Trends in PPPs in Africa

The years 2000 to 2005 saw a steady growth in both the number and value of PPP projects, from 29 projects costing US$3.5 billion to 42 projects costing US$8.9 billion. There was a sharp drop in the number of projects from 2005 and especially after the 2007–2008 financial crisis. The number of projects in recent years has hovered between 16 and 18, and total investment, which peaked at US$13.6 billion in 2008, has fallen to around US$11.4 billion (World Bank 2013a). This means that the average size of project has increased sharply, from US$212 million in 2005 to US$633 million in 2011.

The legal and infrastructural provision for PPPs in Africa

The World Bank lists only 16 African countries as having legislation dealing with PPPs.² The coverage of the legislation varies greatly from country to country, with South Africa having perhaps the most sophisticated legal and institutional structure. Interest in PPPs in South Africa dates back to 1997. A strategic plan for them was endorsed in 1999, a dedicated PPP unit was established in the Treasury in 2000 and legislation was enacted in 2004 for the national and provincial governments and in 2005 for municipalities engaging in PPPs³ (Burger 2006). A PPP manual was issued in 2004 with modules on issues such as procurement practice, negotiating contracts, establishing a public sector comparator against which to evaluate bids, accounting treatment and Black Economic Empowerment (BEE) (South Africa 2004). Recent legislation in Tanzania likewise provides a detailed organisational structure, procedures and requirements for PPPs to be considered as viable options, through value-for-money and other assessments (Tanzania 2010). It is similar in content to that in Zambia (Zambia 2009). Other acts are less detailed and prescriptive, laying out principles of partnership, including tendering, but leaving the organisational framework ambiguous (e.g., Tunisia 2008, cited in EBRD 2011, and Cameroon 2006). In the case of Tunisia, for instance, the European Bank for Reconstruction and Development (EBRD 2011) has concluded that the legislation is too restrictive to allow PPPs to enter the ‘merchant sector’, meaning, health, education, prisons and public buildings. The EBRD also found the policy framework, institutional framework and PPP law enforcement to be quite deficient.

There are, apparently, even fewer countries with specialised PPP units than countries with PPP legislation; in fact, only a total of 10, Angola, Egypt, Ghana, Kenya, Malawi, Mauritius, Nigeria, Rwanda, South Africa and Tanzania, appear to exist.⁴ Some, such as those in Nigeria, Kenya, Rwanda and Tanzania, are quite recent creations, after many years of implementing PPPs.

While having legislation and a specialised PPP technical unit are no guarantee that PPPs will be properly conceived, evaluated and governed, it is unlikely that success will be achieved without them and the majority of African countries pursuing PPPs do not have them. For example, neither the DRC, which has seven PPP projects costing US$2.0 billion, nor Côte d'Ivoire, which has 17 projects costing US$3.3 billion, appears to have a legal or institutional framework for PPPs.

What are the claims made for PPPs?

Proponents argue that there are both macroeconomic and microeconomic benefits of promoting PPPs. On the macro side, they claim that by allowing the private sector to finance projects directly, PPPs help governments to relieve the shortage of domestic funds and reduce their debt obligations. PPPs enable them to reduce their own direct spending or debt, shifting the immediate burden onto the private sector and off the books of the government. When PPPs levy user charges, as they often do with toll roads and the provision of water or electricity, this is said to reduce the pressure on taxation. Reducing the effective size of government through PPPs is consistent with neoliberal emphases on deregulation and economic restructuring and with the priorities, wishes and conditions of aid donors who might offer financial support for PPPs. It is also claimed that PPPs reduce corruption by requiring greater transparency in infrastructure spending decisions.

More important than these arguments in recent years has been a series of microeconomic claims for PPPs. The first is that the private sector is more efficient than the public sector and that it brings projects to completion more cheaply and on time. Involving the private sector in the direct provision of financing introduces a new dimension of capital market discipline in the building and operating of projects, putting added pressure on contractors to deliver promptly and on-budget. Furthermore, private sector involvement can lead to more efficient operations and maintenance relative to conventional approaches. These factors are said to lead to lower life-time costs of projects, from the design stage to the end of the useful life of projects (World Bank 2012, 16).

In short, PPPs shift financial risks from the public sector to the private sector at all stages of a project's life. Indeed, risk transfer is now said to be the most important feature of PPPs. The main risks are project or construction risk, operating risk, market or demand risk, financing risk, environmental risk, regulatory risk, legal/political risk, foreign exchange risk, public policy risk, force majeure and residual value risk (Loxley and Loxley 2010, 35). The biggest risks in most projects are up-front project risk and ongoing market or demand risk. Proponents claim that PPPs shift risk onto those best positioned to handle it.

A final argument for PPPs in South Africa is that they can be important instruments for promoting BEE (Farlam 2005). The South African manual on PPPs contains a code of conduct for BEE which gives numerical weight to furtherance of BEE in PPP submissions and provides for monitoring to avoid tokenism and window dressing (South Africa 2004; Khatleli and Root 2010).

Arguments against using PPPs

In response to the extravagant macro arguments in favour of PPPs, it must be recognised that PPP leases paid to the private sector do not replace debt. They are simply debt in another form as they must be paid, by force of legal contract, over a number of years. The annual lease payments can be given a present value⁵ equivalent to debt, which is exactly what bond houses do when reviewing the financial commitments of governments. Furthermore, PPP financing is almost always more expensive than direct borrowing by governments. Domestic revenue difficulties which it is claimed PPPs help address are, in fact, often self-imposed by governments themselves or externally imposed by the IMF or World Bank, the very organisations promoting PPPs. If PPP lease payments are in reality debt, and are more expensive than direct borrowing, then they really offer no release from domestic revenue and debt difficulties in principle and PPPs must be justified, if at all, on micro grounds. Certainly, PPP funds are not ‘free’ funds and may not even be ‘additional’ funds unless donors insist on their funding being used to promote PPPs.

User fees do not necessarily reduce taxation pressures and may in themselves generate conflict and resistance and revolt, as they have in other parts of the developing world (Marin 2009).

Contrary to the received wisdom of their proponents, PPPs may actually facilitate corruption, as they have done in the energy sector of Tanzania. In this case, a 20-year deal was struck with a PPP to provide electricity that was both not needed and highly overpriced. This was negotiated without a tender and the deal was denounced by the Tanzanian government itself as reeking of corruption (World Bank 2012, 207; Farlam 2005, 28–29). Transparency International called it ‘public–private partnership at its worst’ (Cooksey 2003, quoted in Farlam, 2005, 27). PPPs are clearly not a panacea for corruption and may, in fact, open up new opportunities for it. Indeed, Moeletsi Mbeki has claimed that BEE, being promoted through PPPs in South Africa, ‘is legalised corruption’ serving to ‘co-opt and bribe the controllers of political power, with the elite using taxation and corruption to enrich itself’ (quoted in Taylor 2012).

If there is a case for using PPPs in Africa, then it will have to be based on the micro arguments, but these too are suspect. PPPs are complex, long-term arrangements. They necessarily involve complicated legal documentation and lengthy contract negotiations, which means large up-front costs and delays, often ignored when examining the on-time and on-budget claims for PPPs. Because they transfer control and possibly ownership of public infrastructure to the private sector, PPPs involve large transaction costs and large, ongoing costs of monitoring, relative to conventional projects. Weak institutions and poor governance, characteristic of most African countries, may make PPP arrangements ‘more ineffective in practice than in theory’ (Pessoa 2010, 1) and ‘market solutions may be more expensive and overall economic efficiency may not improve with the use of PPPs’ (Ho and Tsui 2009, 10).

While PPPs usually involve fixed price construction projects, this is no guarantee against cost overruns or project delays, which may be covered in generous bid prices or in the renegotiation of contracts, which is quite common in developing countries. Furthermore, PPP contracts are not always subject to competitive bidding, which is considered to be vital for best practice. This seems to have been the case in the AES-SONEL electricity PPP in Cameroon, with Demuijnck and Ngnodjom (2011) admitting that ‘We have no idea of how the initial negotiations between AES and the Cameroonian government have come about, i.e., of whether or not there has been corruption in this deal’ (259). In reality, for very large projects, there are often only a small number of qualified bidders accompanied by a small coterie of international consultant firms. Once a contract is negotiated, the government is at a competitive disadvantage if the private partner needs or insists on contract renegotiation as this is completely outside the competitive process.

Even when competitive processes are followed, PPPs are often subject to severe information asymmetry, with the private partner having much more knowledge and experience than the public partner, which makes for very uneven bargaining power. One can expect that asymmetry to be large in Africa.

PPPs might even reduce the transparency around the provision of public services in the dozen or so African countries currently having freedom of information provisions in their laws or constitutions (http://www.freedominfo.org/regions/africa), because what were previously public sector activities, possibly open to information requests, now become subject to commercial secrecy. Even in countries with fairly developed information systems, it is very difficult to follow PPP activities and once the high risk construction phase is past, ownership of PPPs is often ‘flipped’ to others, sometimes to off-shore tax havens, with almost no publicity (Loxley 2012). For the most part, however, access to information is a general problem in Africa regardless of PPPs.

The very long-term nature of many PPPs introduces new forms of inflexibility into public infrastructure. Thus, while the demand for public services offered by PPPs may change, the contracts do not, committing the government to continue paying for a facility or service which is no longer needed or at least, not in the form it was previously negotiated. For example, in the UK, PPP schools have remained a financial burden on the public sector long after the need for them has disappeared with demographic changes (Loxley 2012).

The greater efficiency and reduced costs that are claimed for PPPs may simply be at the expense of labour. Replacing unionised workers with non-unionised workers or reducing the wages, benefits and terms of service of workers previously hired by government can often be a major attraction of PPPs, but not one that endears itself to workers.

Finally, while the assumption of risk transfer from the public to the private sector is crucial to the micro arguments in favour of PPPs, this is not an established fact in practice. Project delays and cost overruns, perhaps the biggest sources of risk, are common problems in Africa and are well documented. They can be traced back to ‘problems in finance and payment arrangements, poor contract management, material shortages, changes in site conditions, design changes, mistakes and discrepancies in contract documents, mistakes during constructions, price fluctuations, inaccurate estimating, delays, additional work, shortening of contract periods, and fraudulent practices and kickbacks’ (Mansfield, Ugwu, and Doran 1994). They may be due to ‘increase in the cost of construction materials, poor planning and coordination, change orders due to enhancement required by clients, excess quantity during construction (Nega 2008, vi) and possibly to changes in exchange rates, inappropriate contractors and force majeure (Nega 2008, 50). It is not clear what PPPs might have to offer to address the problems identified if African governments are corrupt, face soft budget constraints or allow contract renegotiations. As for demand risk, private partners often negotiate revenue guarantees from government, as has happened with toll road PPPs in South Africa. Thus, the N4 toll road from Witbank South Africa to Maputo, Mozambique, a 30-year PPP costing Rand 3 billion, would not have happened without guarantees from both governments of both debt and possibly equity (Farlam 2005). Demand or revenue guarantees were also necessary for both the Chapman's Peak toll road in Cape Town (ZAR0.45 million) and the much larger Gautrain in Johannesburg (ZAR23 billion). Indeed, it has been suggested that without government guarantees, PPPs are unlikely to be successful in Africa (Farlam 2005, 35). What this means is that substantial risk continues to be faced by the public sector rather than being transferred to the private sector.

One may conclude, therefore, that despite claims to the contrary, PPPs are not necessarily more efficient than traditional procurement and one must be careful in assessing these claims.

A brief survey of PPP performance in Africa

The problem with evaluating performance of PPPs in Africa is that so little is known about the details of individual projects. Without access to contracts and detailed value-for-money calculations, it is impossible to properly evaluate projects. Even the most sophisticated PPP unit in Africa, that of South Africa, does not publish such details. All that can be found on its website is a list of pending projects with very little information such as sector, ministry and status. There is more information for signed projects by sector, sponsor, financing structure, type, duration, private partner, BEE arrangement, project advisors and net present value to government of benefits and costs. There is, however, no detailed information on how net present value was arrived at or the risk assumptions underlying it, and no details of contracts, financial agreements and schedules. Project reviews go no further than general observations or issues that might hit the newspaper (e.g. Farlam 2005 or The Trade Beat 2012). It is, therefore, impossible to say how well PPPs perform in Africa beyond very general observations.

In that spirit, it can be said that PPPs in the telecoms sector appear to be the most successful. The main reason for this is that new technologies promoted by the private sector produced large productivity gains in the broader context of deregulation. These were generally so high that the private sector earned significant profits, consumers obtained much better service and at lower rates and governments benefited from increases in taxation, an unusual win–win situation all round (Auriol and Blanc 2007, 6). In countries such as Rwanda, Uganda, Ghana and Nigeria, PPP arrangements were used to subsidise expansion of telecom systems into more remote areas (Williams and Falch 2012). Little wonder that this is the most prominent sector for PPPs in Africa, though the scope for further growth of PPPs is probably now quite limited.

PPPs in the water and electricity sectors have been much more problematic. Part of the reason is that providing services to the poor can be very difficult. The poor often live in areas which are difficult to access and costly to service, have ambiguous or no title to the land they live on, may not be able to afford services at full cost, and sometimes have a payment culture which might pose a high risk to the private partner. Because electricity and water, in particular, were heavily subsidised by governments in the 1990s, by 40% and 70% respectively (Harris 2003, 9), the introduction of PPPs as part of a broader neoliberal political agenda was often associated with large increases in the cost of service and this sometimes led to dramatic public protests. This was certainly the case in South Africa where thousands of protests took place over rising costs of water,⁶ sewage and electricity, accompanied by falling consumption, rising non-payment of bills and rising rates of disconnection of services (Bond 2010). In some cases where the rates were actually reduced by the introduction of PPPs, this was only possible because the public sector had deliberately raised tariffs significantly (excessively?) prior to the PPP being formed (e.g. this occurred in Gabon and Côte d'Ivoire, see Marin 2009, 109–112).

It seems that PPPs have been successful in raising ‘efficiency’ in the water service area, largely by reducing water losses, by raising collection rates and by laying off workers, (Marin 2009, 116). PPPs have also had some success in providing water to rich communities in Africa, but have been less successful in providing water services to the poor (Farlam 2005). It seems that provision of water to the poor requires a high degree of subsidisation, especially on the connectivity side, as has been the case in Côte d'Ivoire, Senegal, Cameroon and Morocco, (Marin 2009, 136) and is clearly politically very sensitive. Little wonder then, that private investment in water PPPs in Africa has virtually dried up!

Road PPPs can also be politically sensitive, though less so if alternate routes are available. In Nigeria, there were protests over the opening of the Lekki Toll Road Concession on the Lekki-Epe Expressway. The concession is a (PPP) scheme between the Lagos State government and the Lekki Concessioning Company and involved the rebuilding of a previously federal road. In the process, transport fares rose by 50%. The protests turned violent when police attacked protesters. (News Agency of Nigeria 2011).

There have been major protests in South Africa over road tolls on the Gauteng Freeway Improvement Project but this is not a PPP. There are three road concessions in South Africa, each funded by tolling, but these do not appear to have generated protests. Given that South Africa seems to be quite capable of building roads through the state-owned South African National Roads Agency Limited (SANRAL) with or without tolling arrangements, one has to wonder why PPP road concessions are granted at all in that country.

There are several hospital PPPs in South Africa, but they are too recent to review their performance record and there is insufficient information available to assess how they compare with publicly delivered hospital services. The record elsewhere of hospital PPPs, including Canada and the UK, leaves much to be desired in terms of cost effectiveness, quality and value for money (Loxley and Loxley 2010, 106–111).

The World Bank estimates that 48 projects in SSA, accounting for 11% of total PPPs and 5% of the total investment in PPPs, have either been cancelled or are in distress (World Bank 2013a). What is needed is a careful evaluation of the rest.

Conclusion

We would conclude that African governments should be cautious in the use of PPPs, for while they are being promoted as being self-evidently beneficial, the case for using them is far from convincing. There is a need for governments to have clear policies in place as well as a legal framework and technical capacity to evaluate the value for money of PPPs relative to traditional procurement methods. This might at least enable them to avoid egregious mistakes, the full cost of which is likely to be apparent only after many years. Likewise, governments should be cautious of international institutions, such as the AfDB, AU, UNECA etc., which uncritically promote PPPs. Governments that share this scepticism of PPPs would be better placed to resist further encroachment of private capital on public sector activities if they focused first on improving public sector efficiency and on raising local tax capacity.

Note on contributor

John Loxley is a Professor of Economics at the University of Manitoba, Winnipeg, Canada. He has advised a number of governments in both Africa and Canada on issues of public finance.