Determinants of the successful signing of public–private partnership contracts for water and sanitation infrastructure financing in Zimbabwe

Infrastructure is an indispensable component of the Sustainable Development Goals (SDGs), given its cross-cutting relevance across the 17 SDGs (Renwick et al. 2018). Although the impact that infrastructure has on the SDGs can be either direct or indirect, sector-wise, SDG 7 (affordable clean energy) and SDG 6 (clean water and sanitation) have the broadest direct influence on the SDGs (Thacker et al. 2019). Practitioners and academics have linked SDG 6 to SDG 3 (health), SDG 4, (education), and SDG 11 (sustainable cities) (Mugagga & Nabaasa 2016; Shah 2016). It can thus be argued that SDG 6 is an instrumental multiplier goal with the capacity to decide Zimbabwe's development pathway, as water and sanitation are at the centre of socioeconomic development. SDG 17, target 17.16, and target 17.17 recommend partnerships that mobilise and share knowledge, expertise, technology, and financial resources to support the achievement of SDGs (United Nations [UN] 2023). In Africa, public–private partnerships (PPPs) have received wider acceptance as several countries are reported to have projects in the pipeline (African Development Bank [AfDB] 2020). Nevertheless, there is no unanimity between academics, practitioners, and governments on what constitutes a PPP (OECD 2012). Organisations differ in their conceptualisation of PPPs and definitions that emphasise various dimensions of PPP agreements have been promulgated (Jomo et al. 2016).

The UK Home Treasury (1998), accentuating inter-organisational relationships, joint investment, and shared objectives, defines PPPs as follows: ‘An arrangement between two or more entities that enables them to work cooperatively towards shared or compatible objectives and in which there is some degree of shared authority and responsibility, a joint investment of resources, shared risk-taking, and mutual benefit’. Partnerships British Columbia (2003), underscoring risk sharing and contractual governance between PPP stakeholders in delivering infrastructure projects, defined PPPs as: ‘A legally-binding contract between government and business for the provision of assets and the delivery of services that allocates responsibilities and business risks among the various partners'. The United Nations Economic Commission for Africa [UNECA] (2005) stresses the importance of private sector resources and technical capabilities, and thus defines PPP as ‘the combination of a public need with private capacity and resources to create a market opportunity through which the public need is met, and a profit is made’. The World Bank Institute (2012) views PPPs as ‘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’. Much like Partnerships British Columbia (2003), the World Bank Institute's definition underpins the long-term nature of the contractual agreement and the assumption of major risks by the private partner. Thus, PPPs exploit synergies under joint innovative use of scarce resources, and managerial expertise where this cannot be attained to the same extent without partnering (Jomo et al. 2016; Lember et al. 2019). The Zimbabwe Investment Development Agency (ZIDA) Act of 2019 considers a PPP to be ‘an agreement between the contracting authority and the counterpart where the counterparty undertakes to perform a contracting authority's function on behalf of the contracting authority for a specified period and the counterpart receives economic benefits for the services rendered’. This definition accentuates the principal-agent relationship in public–private infrastructure development collaboration (Ministry of Finance and Economic Development 2019; Mundonde & Makoni 2023). The ZIDA Act of 2019 stipulates that the private partner is compensated for the risk assumed through mechanics that include parliamentary grants, user charges, and cash flows from the project depending on the contract signed. This variation in PPP definitions makes it imprudent to generalise findings on PPP studies, hence the importance of country- and sector-specific research. We therefore focus our study on the nature of agreements, or contracts, which give rise to the PPPs in Zimbabwe, by specifically considering how they are decided upon.

There are various types of contracts used to formalise PPPs. In the water and sanitation sector, infrastructure PPPs are structured either as concessions, leases or affermage, management or service contracts. Concession contracts are long-term in nature, with contract periods of 20 or more years. Thus, a concession is best suited for infrastructure projects undertaken in countries with strong and predictable economic fundamentals (World Bank Institute 2012). From a public standpoint, concession contracts facilitate the intermediation of private capital for either the rehabilitation or construction of new infrastructure. Due to the long-term nature of the contract, the concessionaire is incentivised to be efficient in the construction, operation, and management of the project. The Asian Development Bank (2008) states that efficiency and effectiveness gains translate into higher economic profit for the operator. Transfer of the full package of financing, construction, and operation responsibilities enables the concessionaire to prioritize and innovate as it deems most effective. Variants of the concession contract such as the Build-Operate-Transfer (BOT) have been proposed (World Bank Institute 2012). The private investor undertakes the construction and operation of a project with ownership of the facility transferring to the public at the expiry of the contract. In turn, the public agency contractually commits to purchasing a minimum level of production in order to guarantee a minimum level of demand that allows the investor to realise the return on investment (World Bank Institute 2012).

Though similar to the BOT, the private investor retains ownership of the asset under a Build-Operate-Own contract (World Bank Institute 2012). The Build-Own-Operate-Transfer provides for private ownership, operation, and transfer of the asset to the public agency. With minimal government interference, the private investor can make economic decisions that include setting prices at levels that compensate for the risk assumed (Ameyaw et al. 2017). With a Build-Lease-Transfer (BLT), the private sector invests in the construction of the project, and once complete, operational obligation is moved to the government under a binding lease agreement that is negotiated beforehand. Through the lease, cash inflows from the public agency, payable over the term of the lease, the private players can recoup costs and earn a return on invested capital (Mutandwa & Zinyama 2015). Unlike a BLT, ownership is transferred immediately when construction is completed in a Build-Transfer-Operate. The private player is then granted the legal right to operate the facility on behalf of the government with the revenues shared as per the agreed matrix (Maposa & Munanga 2021). When the infrastructure facility exists but is in a dilapidated state, a Rehabilitate-Operate-Transfer contract can be used. The private partner commits to refurbishing and maintaining the facility, in return for the right to operate it over the tenure of the contract, after which the facility is transferred to the public sector (Maposa & Munanga 2021).

Other than concession contracts, the ZIDA Act recognises other forms of PPP such as affermage contracts where the private partner is permitted by the public authority to maintain and operate the publicly owned infrastructure (ZIDA Act 2019). Sometimes, authority is granted to the private sector investor to directly bill and collect remittances for the services provided. Although the private operator provides services at their cost, the public sector remains liable for any new investments. The private sector is solely obligated to provide the service, while ownership of the infrastructure assets remains public. On average, the contract tenure ranges from 10 to 15 years, and as such the risk profile is medium, relative to concession contracts (Nwangwu 2018). On the other hand, with a management contract, only the day-to-day management and control of the services provision mechanics are delegated to the private operator. The ultimate obligation for service provision remains the responsibility of the public authority. Normally, the tenure of management contracts is short, ranging from 3 to 5 years, with a limited transfer of financial and commercial risks and responsibilities to the private operator. Management contracts are advantageous in that the public partner benefits from private sector operational efficiencies, while retaining ownership of the infrastructure asset (Nwangwu 2018). A service contract is a performance-oriented contract, anchored on commercial and financial administration that is awarded to a private operator. The private partner receives a pre-determined fee, payable either as a one-time fee payment or any other basis, as stipulated in the terms of the contract. A service contract is ideal when the government is reluctant to relinquish control of public water infrastructure systems to private companies. The effectiveness of service contracts is dependent on the clear definition of the services in the contract design (World Bank Institute 2012). Swai et al. (2018) recommended that the nature of a project to be undertaken must inform the choice of PPP model to be adopted. Stakeholders in the PPP deal must agree on the most effective model applicable to the prevailing infrastructure development scenario. The AfDB (2019) thus recommended that, in Zimbabwe, PPP models and contracts should be tailor made to specific infrastructure sectors and projects to be undertaken.

Accomplishing water and sanitation targets in developing countries is a momentous task that is constrained by multi-faceted challenges (Herrera 2019). In Zimbabwe, water and sanitation projects have stalled due to financial constraints and other governance factors (Zhou & Chilunjika 2018). Jomo et al. (2016) argued that PPPs are a critical source of infrastructure finance if SDG 6 targets are to be achieved. Cognisant of this, the Government of Zimbabwe is making efforts to encourage private participation in water and sanitation infrastructure projects (Ministry of Finance and Economic Development 2019). This study thus seeks to determine the key drivers behind the successful signing of water and sanitation PPP contracts in Zimbabwe. This is pertinent given that largely because of information asymmetry, even in developed countries, there remains a hesitance to participate by the private sector and foreign investors with regard to water and sanitation PPP investments (Panayiotou 2017). Furthermore, the AfDB (2019) suggests that PPP financing policy has to be aligned to a specific country and to infrastructure segments as opposed to a general one-size-fits-all approach. Empirical studies that adopted the AfDB's recommended view are limited for Zimbabwe's water and sanitation sector, a gap that this study seeks to contribute towards in terms of both scholarly and practical insights by confirming the key drivers behind a country's ability to enter into and sign PPP contracts for the water and sanitation sector.

In order to fulfil its objectives, this study uses secondary data on water and sanitation PPPs in Zimbabwe for a period of 25 years extending from 1996 to 2021, as extracted from the World Bank's Private Participation in Infrastructure (PPI) database. According to the databank, water and sanitation projects comprise water generation and distribution, along with sewage collection and treatment. Since the PPI databank does not provide comprehensive coverage of small water and sanitation projects (Jensen & Blanc-Brude 2006), the Reserve Bank of Zimbabwe (RBZ), World Development Indicators, and the World Governance Indicators' databases and the Government of Zimbabwe publications will supplement the PPI databank. Following the IMF (2006), Taguchi & Sunouchi (2019), and Pan et al. (2020), the dependent variable is the count of PPP (NPPP) that reached financial closure. NPPP, the discrete count dependent variable, thus ranges between zero and the maximum number of PPPs recorded per year over the sample period. If there is no water and sanitation PPP signed in a particular year, the dependent variable is zero. As alluded to in the study by IMF (2006), a zero recording may indicate the absence of new financial flows into the water and sanitation sector through the avenue of PPPs even if the determinants in this study do not provide an inherent justification for not having a signed PPP in the same year. On the other hand, a high count of PPP deals that reach financial close is an indicator of private sector preparedness and commitment to finance water and sanitation PPPs (IMF 2006). Defining the dependent variable in this way distinguishes this study from others that either used survey strategies or qualitative desktop strategies to investigate PPPs in Zimbabwe (Sharma 2011; Sai et al. 2015; Chitongo 2017). The explanatory variables are summarised in Table 1.

Similar to the approach of Nxumalo & Makoni (2021), this study uses the principal component analysis (PCA) technique to construct a composite institutional governance quality index (GIX). Using a composite index is necessitated by the observation in the literature that the World Bank's governance indicators are highly correlated (Jensen & Blanc-Brude 2006; Nxumalo & Makoni 2021). Moreover, there is no consensus among researchers on which governance indicators exert the most pertinent impact on water and sanitation PPP investments (Jensen & Blanc-Brude 2006; Banerjee et al. 2006). Conducting the PCA analysis requires the estimation of the eigenvalues of the correlation matrix of the original dataset. The components with high eigenvalues summarise the critical information about the original dataset and account for the greatest variation in the dataset (Nxumalo & Makoni 2021). The eigenvalues of the correlation matrix of the six institutional variables are summarised in Table 2. The first component explains the large proportion of the variation in the dataset (83.67%), and the respective eigenvalue is 5.02037.

Table 3 summarises the eigenvector loadings. It is evident that the first component (PC1) has positive coefficients across the six dimensions of measuring the quality of governance. This suggests that the six measures of governance – control of corruption (CC), regulatory quality (RQ), rule of law (RL), voice and accountability (VA), political stability (PS), and government effectiveness (GE) – played a positive role in explaining the overall governance quality in Zimbabwe. It can thus be deduced that PC 1, relative to other components, embodies the critical information with regard to the institutional governance environment in the original dataset.

Having applied a data reduction technique to model the variance structure of the institutional governance variables, and afterwards generate the composite index (GIX), we now specify the regression model that best suits our study.

The average VIF is 2.94, which indicates that multicollinearity is sufficiently managed (Chikaza & Simatele 2021). On the other hand, the use of robust standard errors and the log transformation of explanatory variables controls for heteroskedasticity in the model (Brooks 2008). The findings of the study are presented in the following section.

The research findings of our study are presented in Table 5.

Consistent with earlier studies (Jensen & Blanc-Brude 2006; Sharma 2011; Pan et al. 2020), GDP per capita significantly and positively influences the number of water and sanitation PPPs in Zimbabwe at the 1% level of significance. High and growing GDP per capita is an indicator of an affluent expanding market with ample investment opportunities (Bodie et al. 2013). The opposite holds true. Post the global financial crisis of 2007, countries that are lead adopters of PPP for infrastructure development are characterised by strong GDP per capita (Nikolić et al. 2020). More so, the demand for PPP projects in Europe and selected Latin American countries responded positively to the drop in GDP during the financial crisis (Nikolić et al. 2020). The import cover in Zimbabwe has largely been very low (Kavila & Roux 2016). Contrary to Nakatani (2017), who established that the level of import cover impacts infrastructure investment through the exchange rate channel, in Zimbabwe, the variable was found to be insignificant. Likewise, this study established that the level of inflation does not determine the number of water and sanitation PPPs in Zimbabwe. This finding can be explained by the realisation that inflation risks in water PPPs are hedged against through claims on precious minerals, especially gold. However, Tshehla & Mukudu (2020), through qualitative survey analysis, identified inflation to be a significant determinant of project finance structures in Zimbabwe. We further unearthed a significantly negative relationship between Foreign Direct Investment (FDI) and the count of PPP contracts, implying that the inverse relationship exhibited points to the fact that private investment in water and sanitation is needed but foreign investors are discouraged by the prevailing economic and investment environmental conditions in the country.

In Zimbabwe, financial market development influences the signing of water and sanitation PPP contracts. Consistent with Ba et al. (2017), stock market capitalisation significantly determines the count of PPPs at 1%. The World Bank (2011) noted that, in Africa, stock markets often fail to consistently and adequately finance infrastructure, due to weak market sophistication and shallow services' offerings. Nevertheless, the Zimbabwe Stock Exchange has a long functional history, and strategies have been implemented to enhance the financing potential of the market. This study also confirms that the level of bank market development is a key determinant of the signing of water PPPs. The finding is consistent with Kamau (2016) and Rao (2018) who concluded that, through the credit channel, banks contribute positively to infrastructure development through PPPs. NPLs have a significant and negative causal relationship with the signing of water and sanitation PPPs in Zimbabwe. The finding supports the proposition that non-performing assets curtail financial institutions' lending potential, while high asset quality enhances the propensity to lend towards project finance deals. The RBZ (2020) stated that, since 2015, non-performing loans to total loans ratios have shown improvement, due to the effectiveness of debt recovery efforts, as well as the effective adoption of domestic credit checking processes and infrastructure by most banks in Zimbabwe. The governance composite index is significant (at 5%) and exhibits a negative relationship with PPP infrastructure investment. This confirms that the state and level of institutional governance within the country play a pivotal role in the decision-making of prospective PPP investors, particularly foreigners and the private sector, into water and sanitation projects in the country. Chinese investors, with limited regard for the state of governance in the host country, have become the main sponsors of water PPPs in Zimbabwe, due to limited investor alternatives. Diversifying sponsors and PPP players requires government commitment to improve institutional quality and governance scores, as many foreign investors are sensitive to the institutional governance environment in the project host country. A number of studies have stressed the importance of governance in PPP investments (Banerjee et al. 2006; Taguchi & Sunouchi 2019; Fleta-Asín & Muñoz 2021). A good track record on institutional governance variables strongly encourages the signing of PPP contracts.

This study concerned itself with the identification of the key determinants of the number of water and sanitation infrastructure PPP contracts signed in Zimbabwe during the 1996–2021 period. This study was timely in that Zimbabwe remains under strict economic sanctions from the West (the United States and European Union), but enjoys strong diplomatic and economic ties with the East (China). Despite this, the country remains optimistic that it is attractive to private investors, some of whom are interested in the infrastructural development opportunities that the country presents to them. This study has highlighted the importance of developed financial markets (both bank credit and stock markets), as well as good quality institutions, which are necessary absorptive capacities the country must possess if it stands any chance of harnessing the necessary capital to fund its multiple infrastructural development projects. For the private sector, the contracts must yield a substantial return on investment, or additional value, while for the public sector or government, the project should efficiently (operationally and financially) service the citizens in need. The Zimbabwean government should therefore make efforts to attract more private domestic and foreign investors to finance its water and sanitation infrastructural development, by strengthening its institutional governance and policy framework and allowing the bi-directional flow of capital by easing capital openness.

The Government of Zimbabwe also needs to foster further bank market and capital market development in order to enhance credit availability and investment opportunities for PPP infrastructure projects.

This article's limitations are that it primarily focused on a single developing country, which has been economically isolated for several decades and the variables analysed may not apply to later phases of water and sanitation projects in Zimbabwe. As such, the findings may have little to no generalisability to other developing countries in the region. Hence, although this study contributes to the ongoing debates on water and sanitation infrastructure contracts in general, using Zimbabwe as the unit of analysis; future studies can extend the focus to other developing countries such as Botswana, Mozambique, Malawi, South Africa, and Zambia for comparison purposes, particularly since those countries are not affected by sanctions as is the case with Zimbabwe. Also, our present focus was on the water and infrastructure sector, so future research can consider a cross-sector analysis to include other infrastructures such as transport and Information and Communications Technologies (ICT), as their determinant factors may vary from the recently identified ones pertaining to a single sector.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.