Between 2004 and 2012, investment in public-private partnerships (PPPs) in developing countries increased nearly six-fold, from US$24.4 billion to US$144 billion, after which, there was a dip for two years before it bounced back to US$120.2 billion in 2015 (Romero and Vervynckt, 2017). Such significant increases call for the need to measure the performance and contributions of these investments. In doing so, identifying the factors behind their successes and failures can provide essential lessons for improving the future design and implementation of these partnerships. PPPs are often thought of as projects with social goals and private expertise, such as investments in infrastructure facilities for public goods with universal entitlement, irrespective of the ability to pay. Thus, even more so, social goals need to be kept in mind when defining and measuring their performance.
Pro-poor objective of PPPs
One of the central goals of the World Bank Group is to fight poverty and the PPP initiatives it supports need to be aligned with this strategic social goal. However, PPP projects, in general, tend to focus more on economic development through infrastructure growth rather than on poverty reduction. When an economy grows in parallel with falling poverty rates, economists call it ‘pro-poor growth’. Development experts hold that PPP initiatives must have benefits, outcomes and welfare distribution channels that are pro-poor. For example, PPP performance indicators state that pro-poor PPP projects should display an increase in connectivity or access for poorer areas, include piped water or electricity projects, improve access to health services such as medical clinics, enhance public transport facilities, or contribute to the development of economic zones.
The Independent Evaluation Group (IEG) to the World Bank Group issues an important evaluation report that underpins support for PPPs. Its reports for the 10 years leading up to 2012 provide reviews for PPP interventions around the world, after which they seek to provide recommendations to the World Bank Group, including the International Financial Corporation (IFC) and the Multi-lateral Investment Guarantee Agency (MIGA). Among other tasks, the evaluation teams provide project-level assessments of individual PPP projects.
In its 2015 report, the IEG stated that, under its new strategy, the World Bank Group intends to work with the public and private sectors to end extreme poverty and promote shared prosperity. PPPs financed by the World Bank Group must thus include pro-poor objectives of providing infrastructure or services. Project-level monitoring and evaluation systems must be properly designed to include this target as a measurement for a project’s success.
Pro-poor data scarcity
Different evaluation systems are used around the world to measure the performance of PPP investments. Their focal point is whether or not individual investments are able to record their achievements in addressing the concerns of stakeholders, including poor beneficiaries.
Meanwhile, in its 2014 report titled, ‘World Bank Group Support to Public-Private Partnerships’, the IEG acknowledged that there was a big gap in monitoring and evaluating pro-poor data. The available numbers are mainly financial performance indicators, including those used for cash flow estimation, while there is a paucity of data for other PPP success dimensions such as pro-poor aspects and fiscal effects. Although the central goals of the World Bank are to fight poverty and promote prosperity, PPP internal audit teams have emphasised financial performance as a PPP project success indicator and have seemingly forgotten that PPP projects in developing countries should be for the poor.
Figure 1 shows PPP projects supported by the World Bank, IFC and MIGA, together with data on the available results. Disappointingly, out of a total of 442 PPPs supported across the three organisations, none offer data for all (financial and social) dimensions. Meanwhile, there are less than 10 PPPs, or 8 percent of the total, with available data on pro-poor dimensions. This ratio is too small to determine whether the Bank is meeting its strategic goal of assisting the poor.
Furthermore, in the IEG Report, 22 PPP cases were selected for an in-depth assessment to measure their social performance indicators. These PPPs are located in three regional clusters: Latin America (Brazil, Colombia, Guatemala), East Asia Pacific (China, the Philippines, Vietnam) and Sub-Saharan Africa (Ghana, Senegal, Uganda). This assessment showed that the measures for pro-poor performance indicators for water and transport PPPs (i.e. access for the poor or high coverage in poor areas) are exceptionally high.
A rare pro-poor case study: The Philippines
Among the rare cases of addressing pro-poor objectives, the IEG report noted several initiatives in the Philippines that had benefited the poor. The field of development interventions in the country included improved water connections, electricity connections and transport participation, which required some funds contributed by users. The interesting features of the PPP cases in the Philippines include the fact that the poor not only received subsidised access to PPP infrastructure, but were also presented with other affordable options. For example, minibuses, which are convenient for picking up and dropping off passengers, enjoy a cheaper daily flat rate for using the PPP highway compared to bigger vehicles, which made it more affordable for poor commuters. As a result, the share of minibuses traffic steadily increased from 0.9 percent in 2004 to 7 percent in 2010. For electricity PPP projects, poorer households were given the choice of geothermal, thermal and hydro-power generation, all subsidised by the government, for water supply, a pilot PPP project subsidised the connection fees, which helped to raise connection rates in poorer areas, and this model has been planned for replication (refer to Figure 2).
Challenges with pro-poor projects in Ghana
Implementing a pro-poor PPP project in Ghana revealed a conundrum for the private sector. On the one hand, the private investor was attracted to developing public infrastructure with purely profit incentives and focused on formulating a pricing structure where tariffs were set at a price high enough to recover the initial investment and bring in subsequent positive income. On the other hand, it was under pressure not to increase the financial burden for the poor. In fact, social movements in Ghana such as the NGO Integrated Social Development Centre worked with the Coalition Against the Privatization of Water in Ghana to carry out anti-PPP campaigns. As a consequence, private investor Ghana Company Water Limited had to set a low tariff of 31 cents per cubic metre, much less than its initially proposed 70 cents per cubic metre to financially recover the investment as projected. The income from tariffs collected was so inadequate that it was financially unhealthy and, noted Fuesta and Haffnerb (2007), “it survived only by means of considerable government subsidies.” Meanwhile Lobina and Hall (2003) observed that a private company with a thoroughly profit motive finds it, “…very difficult to reconcile with service delivery to the poor without substantial public subsidy.”
Why is pro-poor data scarce?
Evaluation experts at individual PPPs must have known of the social objective to include the poor. It is not difficult for PPP teams to record data on how many poor households are included in the pool of beneficiary households and are connected with piped water systems or electricity grids funded by the project, and then to include this ratio in evaluation reports. But most of them, as compiled in the IEG report, did not do so. There are several justifications for why this so.
First is the ‘disconnect’ between objectives and practice. While these objectives have been set by the World Bank, in practice, teams struggle with such objectives. For example, rural PPP piped water use entails connection costs and periodic tariffs. Due to the limited budget, local grants may pilot funding for just a small number of connections with a discount in the project’s first year, but not for total costs and monthly tariffs (e.g. the Philippines water PPP case). Poverty coverage is therefore low and may become lower as connected households abandon piped water to avoid monthly tariffs. But evaluators do not want to see a decrease in the number of poor beneficiaries in their reports. One possible solution is to plan tariffs differently for different income groups. For example, the very poor should be exempted from or charged a very low tariff to stimulate water usage.
Second is the issue of being ‘solution-based’ and not ‘problem-based’. Development designers stick to a solution, which is possibly successful in one context and want to replicate it in all contexts. In fact, each context is characterised by individual complexities that require a matching solution. Teams may come to a locality with the idea of addressing the ‘lack of tap (piped) water’, for example. The truth is that ‘tap water’ is not a problem but is just one of several options used to resolve the problem of health or more specifically, water-borne diseases. Practitioners can select the best option to address this problem in a way that is affordable, feasible and the most impactful for local people.
For example, sparsely populated areas are normally not financially and technologically suitable for building a water pipeline. If the policy or water use master plan is to popularise tap water everywhere, including sparsely populated areas, on the pretext of modernising rural locations, connection rates may not be high as expected, as financially struggling households may use their limited resources for other priorities, such as their children’s education or healthcare services. More evidence is needed to support the relationship between the use of tap water and reduction in occurrence of water-borne diseases, such as diarrhoea. It is generally claimed that diarrhoea is also due to unhygienic eating and drinking behaviour.
Third, in the PPP implementation process, there may be stakeholders that are considered more important than the poor beneficiaries. For instance, private investors are very interested in the profitability of a PPP project. Specifically, they would like to know how many households are willing to pay for connections and tariffs so that they know when a project would pay back their initial investment capital. They may be concerned if there are many households without the capacity to pay. Freeman and McVea (1984) suggest a two-variable framework consisting of stakeholders’ interests and power, which holds that a stakeholder with strong authority or power and a high level of interest in the project outcomes requires them “to [be] manage[d] closely”.
Private investors have a strong voice because their decision determines whether or not the project is funded. Therefore, PPP projects may struggle or may not want to report the number of households without the capacity to pay, unless there are guaranteed sources of grants for them (as in the case of the piped water PPP in the Philippines). This hesitation may lead to the exclusion of pro-poor indicators, which means evaluation reports will miss this data.
In summary, it is conventionally believed that public infrastructure facilities are public goods rather than commercial goods and everyone is entitled to use them. The World Bank Group strategy states that PPP beneficiaries should include the poor who have difficulty paying for those services. This presents a potential conflict of interest with private investors who seek financial returns from users. Even so, several options are available to tackle this challenge. Private companies may follow corporate social responsibility (CSR) practices including subsidising and providing grants to the poor. The design of monitoring and evaluation systems can also allow for some flexibility in recording pro-poor data, depending on local conditions. Sources of pro-poor data may also come from private sector CSR reports, not necessarily from project records. Nevertheless, any PPP will need to set up a monitoring and evaluation system to make a comprehensive assessment of PPP project success, including the pro-poor aspects. Without them, PPP results will not be completely aligned with the World Bank Group’s strategy for the poor, nor with the many NGOs that keep asking the Bank to meet this social objective.