“Look Before You Leap”: Lessons from Urban Water Sector Reforms in Ghana

Introduction

According to their proponents, management contracts (hereafter, MCs) generate greater value for money (VFM) than traditional bureaucratic means of public service delivery. MCs, accordingly, are held by some to be a panacea for the problems of providing critical social services in developing countries faced with weak and ineffective public organizations (Ghandan and Eberhard, 2007; McGuire, 2004). It has therefore been argued that policymakers in developing countries should embrace contracts as a way of infusing critical expertise and private sector skills into the public sector. MCs, with their strict performance targets and enforcement mechanisms, it is held, also ensure greater transparency and accountability in improving service delivery (Ghandan and Eberhard, 2007; Osborne and Waterston, 1994). Nevertheless, many MC projects do not seem to deliver the VFM results that they promise. Opponents argue that explicit contracting cannot address political, cultural and social concerns associated with public organizations that deliver critical social services such as water supply (Lane, 1999). Others argue that the for-profit imperative of private firms means that the democratic process is often circumvented in the name of increased efficiency (Johnston and Romzek, 2004). Despite these arguments, MCs have become important reform tools in the state-owned enterprise (SOE) sector in developing countries like Ghana partly due to the unpopularity of, and opposition to, the privatization of SOEs (Amenga-Etego, 2004; Danso, 2008; Tangri, 1991).

Focusing on the literature and theory on contracting in the public sector, this article is concerned with examining, both empirically and theoretically, the use of an MC for the reform and restructuring of urban water services in Ghana. Two main questions inspire this analysis: (1) is contractualism suitable for the reform and restructuring of critical social services like urban water sector activities in developing countries; (2) under what conditions can contracting actually promote efficiency and effectiveness in such social sectors? It will be argued here that, first, contractualism is flawed both as a concept and as an approach to the reform and management of politically and socially sensitive social services such as water. And, second, for contractualism to actually promote efficiency and effectiveness as its proponents claim, certain local conditions − such as context, agency task specificity and creative adaptation − must be adequately addressed. Unless these conditions are addressed and market-type reforms such as MCs are crafted to suit the politico-administrative context of adopting countries and policy sectors, the policy objectives of improving service delivery will be a mirage.

The analysis is structured as follows. First, it reviews the literature on the use of MCs in the delivery of social services. Second, it provides a brief historical perspective of the water sector in Ghana in order both to place the discussion in a proper historical perspective and to serve as a reference point on which to base the analysis. With these preliminary matters in place, sections three and four conduct post-reform VFM and democratic governance performance evaluations of the MC in the urban water sector in Ghana. The final sections contain a discussion of the findings and a summary and conclusion. Throughout the article, management contracts and contractualism refer to the same concept and will be used interchangeably.

Contractualism as a Reform and Governance Tool: A Theoretical Conceptualization

Ideologically, contractualism is a key component of the neo-liberal idea of shrinking the size of the administrative state by transferring critical sectors of the public sector to private agents to manage (Clark, 2002; Gregory and Christensen, 2004; McGuire, 2004; Romzek and Johnston, 1999). Theoretically, it is also a critical aspect of the New Public Management (NPM) and the reinventing government concepts of “letting the managers manage” and “rowing not steering” that call for a clear structural separation of policymaking and implementation to enhance effectiveness (Borins, 1995; Kernaghan, Marson and Borins, 2000; Osborne and Gaebler, 1993). Furthermore, contractualism originates from the stables of the new institutional economic views such as principal−agent theory that attempt to directly infuse economic theories and principles into political problems in public administration (Dollery 2009; Jones and Butler 1992; Wallis, Dollery and McLoughlin, 2007). Finally, contractualism is a core component of attempts by scholars to introduce new forms of alternative service delivery (ASD) mechanisms into the public sector to help improve service delivery efficiency and value for money (Slyke, 2003; Warner, 2008; Warner and Hefetz, 2008). The question, then, is: whether as an ASD mechanism or as a principal−agent arrangement, what exactly is contractualism in the public sector?

Lane (1999) defines contractualism as a situation where public agencies, policy sectors, personnel and other resources of the public sector are to be managed by private agents through a series of contracts. He argues that such contracts “will cover not only the employment relationship but also be used for the clarification of objectives and tasks for service delivery” (Lane, 1999: 180). From this perspective, contractualism serves as a crucial tool for the structural separation of SOEs where contracts will provide the road map for competition in private markets. The primary reason for the increasing adoption of contracting is the belief that it obviates the rules, processes and red tape of bureaucrats and instead maximizes the expertise of private agents and firms in public service delivery, thus leading to a parsimonious utilization of public resources through the infusion of private sector ideas (Osborne and Waterston, 1994).

Contractualism will be used in this study to refer to a long-term (at least five years) financial, managerial and organizational arrangement between an SOE and a private firm in which certain functions of the SOE are transferred to the private firm for management purposes, usually for a fixed period of time and for a guaranteed fee (Ghandan and Eberhard, 2007; Hubbard, Delay and Devas, 1999). Normally, the contract contains provisions for its extension or cancellation at the end of the stated period, usually based on the attainment of certain predetermined and agreed performance targets. These targets are set in order to ensure that managers of these SOE focus on results and outcomes, not merely inputs and processes, a key condition of the NPM. In this study, the attainment of these performance targets will be used to determine the success or otherwise of the contractual relationship and reform effort in the SOE sector.

Osborne and Waterston (1994) have identified the main elements of contractualism as: (a) an arrangement made for the production and delivery of a specified output; and (b) an amount to be paid for the services being agreed on with the client prior to the start of the contract; which is (c) legally enforceable against the client or contractor under the contract. These elements are expected to reduce the role and impact of the state in the direct management of SOEs and provision of public services. Further, they speak to another fundamental principle and accept an uncontested assumption that governments have become overextended and need to shrink in size through contracting out some services to the private sector. The fundamental features of a contract that separate it from other public policy tools are, therefore, that it is very specific; involves payment for the benefits received; and contains an enforceability clause, including an obligation to pay compensation for contractual failures (Lane, 1999). It can therefore be said that contractualism thrives on the expectations that government programs and services can be very specific, with outputs and outcomes clearly identified and measurable, and with agreed penalties for failure (Johnston and Romzek, 2004). This is not always so.

In addition, contractualism fundamentally reshapes the relationship between citizens and their government along strictly defined legalistic, business and contractual lines, just like a formal arrangement between a private firm and its clients (Osborne and Waterston, 1994). It replaces the implicit relationship between citizens and the state with explicit contracts and performance targets. Such a fundamental reshaping of the relationship between citizens and the state could be highly problematic, because even though citizens want better and improved services from the state, their desire is even stronger to be treated as citizens who have a voice, respect and special rights protected and defended by the state (Bovaird and Löffler 2009; Verhoest, Verschuere, Meyers, et al. 2010). Moreover, and in order to be effective, a “contract needs to specify delivery (what, when, how and by whom − technical agreement) and payment (how much, when, by whom and with what guarantees − the commercial agreement)” (Hubbard et al. 1999: 157) between the principal and the agent. These requirements, however, can be challenging because in the public sector it is difficult to delineate the boundaries between technical and commercial agreements, especially in the provision of vital social services such as water supply that have cultural, social, financial, and even religious functions and values. It can therefore be extrapolated from this theory that contractualism translates political questions into managerialist concerns of the ability to measure and quantify empirical targets, and accountability for clearly stated outcomes.

An inherent problem with such a narrow and reductionist VFM approach, implied in MCs, is that political commitment and sensitivity − which are critical to the success of reforms in the public sector − are difficult to measure or quantify in a contractual situation. This difficulty poses managerial challenges to the successful implementation of contracts in the water sector because of the duality of the tangible and intangible objectives of these contracts. They also mean that MCs are very sensitive to the politico-administrative context of the adopting country. Furthermore, Lane (1999) raises two fundamental questions: whether contractualism “is suitable for all kinds of public sector activities, as well as under what conditions contracting actually promotes efficiency” (Lane, 1999: 180). These questions speak to concerns about the suitability and applicability of contractualism and its ability to answer critical political and social questions associated with the task specificity of different SOEs in different political systems.

Unfortunately, the verdict on MCs in the public sector is not yet in. As Johnson and Romzek (2004) explain, “advocates see benefits such as increased efficiency, reduced size of government, greater program flexibility and more transparent accountability” (Johnston and Romzek, 2004: 108) under contractualism. Opponents, by contrast, “bemoan contracting reforms as misdirected − either as prone to abuse by contractors, or thinly veiled excuses to reduce government and diminish the capacity of government” (Johnson and Romzek, 2004: 108) to provide critical services to the most vulnerable in society.

But despite the controversy about contractualism, it has become a common approach in both developed and developing countries that seek to inject efficacy and increased performance into their public sectors and service delivery mechanisms. It can also be inferred from the foregoing review that contractualism comes with a “buyer beware” label attached to it. This means that its professed benefits are not guaranteed but highly dependent upon the government’s ability to negotiate a good contract and to regulate and monitor the parties to the contract in order to reduce, if not eliminate, opportunistic behaviours and cheating (Dollery, 2009). This means that MCs need to be creatively adapted to suit individual policy sectors and SOEs. It also means that MCs are very sensitive to the management culture of certain sectors and SOEs.

An important lesson to learn from the above literature review is that MCs are complex approaches/tools for reforming SOEs. It is precisely because of this complexity that we need to develop a framework that will serve as a basis for analyzing MC theory in developing countries in general and in the water sector in particular. According to Lindquist (2010), a framework should show how different factors and elements relate to each other and also allow a debate on which norms, values, and even ideas to project and why.

Based on the above theoretical review of MCs and the Lindquist (2010) concept, the undergirding principles of this proposed framework can be distilled and identified as: context, creative adaptation, and task specificity. Context means the sensitivity of reform models/approaches to the needs of specific politico-administrative cultures and systems (Pollitt and Bouckaert, 2012). Task specificity as used here refers to the appropriateness or suitability of MCs in the water sector specifically. Finally, creative adaptation means that MCs must be creatively adapted to fit and suit the managerial, administrative and operational cultures of individual SOEs. Together, these (context, task specificity, creative adaptation) represent the most important factors to be considered for the successful adoption and implementation of MCs especially in the water sector. This framework is apt for the discussion here and will be used as the basis for analyzing the water sector in Ghana because it has the analytical advantage of capturing the ambiguous and multifaceted nature of MCs as public policy instruments with empirical VFM objectives.

Methodology

The study is primarily based on a case study of urban water sector reforms in Ghana undertaken by an MC. This case was chosen for several reasons: the strategic significance of the politics and policy of water reforms; the high profile nature of the case; the amount of money involved; and the availability of data. Qualitative data were collected using specialized face-to-face interviews in Ghana. This approach involves directly engaging and talking to respondents in the interview process to get their input (McNabb, 2010; Singleton Jr. and Straits, 2010). The interviewees were recruited by mailing letters to the leadership and senior management of various institutions and stakeholders in Ghana’s water sector inviting them to participate in the study. The actual interview process deployed involved 12 people in nine different organizations using semi-structured and open-ended questions. These specialized interviews were supplemented with empirical quantitative data that address issues of organizational VFM performance. Quantitative data consisted of the annual audited accounts, quarterly reports, budget statements, performance reports/audits and external performance evaluation reports of Aqua Vitens Rand Ltd (AVRL), the water supply consortium created in 2005. Both qualitative and quantitative data were used in order to minimize bias and to ensure the objectivity and reliability of the information collected. The discussion will now turn to a brief survey of the history of urban water sector reforms in Ghana.

A Brief Reform History of Urban Water Reforms in Ghana

The development of urban water supply in Ghana began in 1928, when the then colonial government constructed the first pipe-borne water system at the colonial capital, Cape Coast. With the subsequent establishment of the Department of Rural Water in 1948, the piped water system was extended to other regional capitals in the colony. Later, in 1958, a Water Supply Division was established to assume combined responsibility for both rural and urban water supply. In 1965, the Ghana Water and Sewage Corporation (GWSC) was created to manage the supply of potable water and sewage systems in Ghana in both rural and urban centres. The scope of the GWSC’s works quickly expanded from 35 pipe systems to nearly 208 by 1966 and then over 1000 by the early 1990s.

In the attempts to pipe water to as many people as possible, supply was skewed toward urban centres. Statistics indicate that by 2002, access to treated water was available to 62–70% of the urban population, while 78% of the poor in the urban areas did not have piped water (IFFM, 2002), with the result that a significant portion of the population in both the urban slums and rural areas were without an adequate supply of drinking water. This discrepancy can be attributed to a number of deep-seated institutional, financial and operational bottlenecks in the sector. These difficulties subsequently led to the initiation and implementation of a number of reforms with support from local and foreign agencies aimed at restructuring the water sector in the 1980s and 1990s. In 1987, under a five-year development plan for the sector, $US125 million was extended to the government by the World Bank and other donors to rehabilitate, renew and expand the water supply systems to meet demand across the country. In addition, between 1990 and 1994, the government, with support from donors, embarked upon a major Water Sector Rehabilitation Project in almost all the urban centres with the intention of addressing institutional, management, operational and maintenance deficiencies (Nkrumah, 2004). But once again, and despite these attempts and the huge sums of money sunk into the system, little progress was made in terms of water supply increases, and access to pipe-borne water continued to be limited even in major urban areas.

In order to reverse these trends, the government, upon the advice of the World Bank (WB), began to look to the private sector for technological, financial and managerial ideas on how to improve the sector in the 1980s (SEC, 1995). This reversal, from public to private sector ideas, must be put in context. It was at the height of the Structural Adjustment Program ¹ (SAP) imposition on the country and the government needed financial and technological resources and advice from the WB to improve the water and other sectors of the economy (Herbst, 1993; SEC, 1995). With the approval of the WB, the government engaged two foreign consultancy firms, Sir William Halcrow & Partners (hereafter, Halcrow) and Louis Berger SA, to study the feasibility and implications of restructuring the water sector and to advise on how to involve the private sector in urban water sector management. The Halcrow report recommended that the government should choose an Enhanced Lease Option together with measures to enable the sector to recover the full cost of its operations through the sale of water (Halcrow, 1995). Based on these recommendations, Louis Berger SA was commissioned by the government and the WB to develop a business plan for the implementation of the lease plan. The Halcrow and the Louise Berger SA reports gave the government ideas for what became known as the private sector participation ² (PSP) process in the urban water sector reforms in Ghana.

An Unholy Partnership? The Tale of “The Grantor” and “The Operator”

The PSP was strongly opposed by a section of Ghanaian civil society and other stakeholders in the water sector who came together to form the National Coalition Against the Privatization of Water (NCAP) (Adam, 2005). The NCAP comprised a cross-section of civil society, labour groups, and International Non-Governmental Organizations in Ghana who were united by their opposition to the PSP (Adam, 2005). An international fact-finding mission on water sector reforms, commissioned by some members of the NCAP stated that “the PSP proposal as it stands is unlikely to improve access to clean water and affordable water and sanitation services” (IFFM, 2002: 8).

Despite these concerns and opposition, the government went ahead and implemented the proposal. It signed a management contract in 2005 between the GWCL, an SOE, and Vitens Rand Water Services BV of Netherlands, a consortium of Vitens International BV of the Royal Netherlands and Rand Water Services Pty of South Africa, which came together to form Aqua Vitens Rand Ltd (AVRL). The contract covered 80 urban water systems, was expected to run for five years with a possibility for extension and was financed by a World Bank grant of US$103 million to the government (World Bank, 2004). Under the proposal, “the ‘Operator’ [AVRL] will be responsible for operating existing and future assets of the ‘Grantor’ (GWCL) in the specified areas, on a day-to-day basis to deliver potable water” (World Bank, 2004: 75). Upon satisfactory performance, the government had the option to enter into a longer-term partnership with the private operator. The final contract sum comprised US$103 million from the World Bank, counter-funding of US$12 million from the Government of Ghana and US$5 million from the Nordic Development Fund, which went directly to the winning bid for what is referred to as minor repairs, rehabilitation and replacement works for the five-year period (GWCL, 2010b). The process to award the contract to AVRL was based on the classical economic theory of open competitive bidding for the best price and service delivery (Prager, 1994).

As well, the 2004 Contract Framework Paper that had been developed by the Government of Ghana and the World Bank to guide the MC recommended the creation of a Revenue Collection Account ³ to be exclusively managed by the Operator and lodged in a commercial bank in Ghana (GWCL, 2004). The Framework Contract specifically stated that “the Operator (AVRL) will be the sole signatory on Revenue Collection Account” (GWCL, 2004: 8). The Revenue Collection Account refers to all incomes and revenue from water sales to both domestic and commercial customers. In other words, the Operator has the complete monopoly and control of any revenue realized in the operations of urban water in Ghana and will decide how to spend this revenue by adopting the appropriate financial and accounting practices it deems fit without any oversight from the Grantor.

The MC was expected to address the pressing financial, managerial, infrastructural and technological weaknesses facing the sector. The plan was to bring on board a private partner with the needed financial, technological and managerial expertise to improve management, operations and performance in the sector. In keeping with the NPM and MC orthodoxy, the aim was to tap into the private sector and private expertise for a more efficient and effective water delivery system in order to improve and increase performance (Barzelay, 1992; Ferlie et al., 1996). Both parties agreed that a decision to extend the contract at the end of the initial five-year period was to be based on the attainment of the specific performance targets by AVRL. The most critical of which was to reduce non-revenue water and improve the finances of the sector. The decision to use this MC in the urban water sector was highly significant in the reform history of Ghana because it marked the first time that the government decided to use a private consortium to manage water in the country. It is therefore obvious that the performance of this MC will have a major impact on the use of similar measures in the future. To determine whether the policy objectives were achieved, we now turn to the post-MC VFM performance analysis of the contract.

Examining the Post-reform Performance of AVRL

The following examination relies largely on quantitative data from departmental annual performance reports, budgetary forecasts and projections, and the annual and quarterly financial results outlined in the initial contract. In 2005, the annual audited accounts of AVRL, presented by an independent external auditor, show that AVRL recorded losses in financial and economic indicators − including the areas of total revenue, billing, collection, operating income (gross profit) and net operating income (deficit) − from a low of 20% to 181% (GWCL, 2006). The company ended the year with a loss of GH¢53 million ⁴ from a projected estimate of GH¢66 million. The statistics were similar for 2006 (GWCL, 2007). An annual performance monitoring and evaluation report by the State Enterprises Commission (SEC) also concluded that the company experienced a loss of GH¢9.5 million from a projection of GH¢11.6 million (SEC, 2006). In 2008, however, the company recorded two positives in net operating surplus and operating surplus (GWCL, 2009a). This positive result, however, is partly due to an upward adjustment of water tariffs in that year and, therefore, had nothing to do with improved management. Yet it was enough to catapult the GWCL into a net operating profit of GH¢32.9 million. Finally, by the end of 2009, the company had recorded a net operating deficit (NOD) of GH¢124 million (GWCL, 2010c) as a result of losses incurred from a shortfall in water sales and other incomes. Thus, even though the company recorded a surplus of GH¢45 million at the end of 2009, it produced a net operating deficit (GWCL, 2009a) due to higher expenditures. Therefore, from a financial perspective, AVRL’s performance can be seen to have been very poor.

Apart from the financial record, AVRL was expected to excel at the drawing of raw water and the treatment, metering and selling of it to customers for a price high enough to make a profit. Under its technical and operational obligations enshrined in the contract, AVRL was specifically tasked to improve substantially the sale of water, the collection of water bills and the metering of water sales, thus increasing the revenue from water sales. These were the areas in which the GWCL was deemed to have no expertise, resulting in the need for a private operator that did. Nevertheless, between 2005 and 2009 the company performed poorly in water sales, production, billing and collection against pre-set targets (GWCL, 2005, 2007, 2008, 2009b; SEC, 2008). Results for 2006 are lacking. ⁵

Moreover, the contract specifically tasked the AVRL with improving service efficiency, and the main component was a demand for the reduction, at an annual rate of about 5%, of non-revenue water. ⁶ This demand must be placed in its proper context in order to better understand its significance. Before the contract, non-revenue water accounted for about 50% of all treated and billable water, which meant that half of the projected revenue and the quantity of water available were lost in transition from the production plant to the final consumer (GWCL, 2010b; SEC, 2009). This loss was due to many causes: pipe leakages, burst pipes, illegal water connections by unscrupulous and criminal elements, and water contractors who siphoned their supply and resold it at higher prices (GWCL, 2005). Improving this aspect of water management was an essential condition of the contract and a major indicator and performance target. Progress in this area would have a large positive effect on the quantity and quality of water available to consumers, because it would lead to more metered and billable water for onward distribution to more people (GWCL, 2009b). That, in turn, would translate into more people getting access to an improved and secured source of water and would produce a further reduction in water prices.

However, there was little or no improvement in reducing non-revenue water. The overall financial and economic performance can therefore be regarded as very poor. This means that AVRL failed to meet its targets for new water connections. Since 2005, the GWCL/AVRL set a specific target for new connections. Surprisingly, the company consistently achieved poor results even in this area. Once again, this outcome can be traced to the chronic inability of the company to achieve its target of reducing non-revenue water.

In conclusion, tangible and empirical data demonstrate that the benefits and results of efficiency, effectiveness and economy that were expected under the management contract did not materialize. Review of the technical, financial and commercial audits of AVRL (the performance targets outlined in the MC) by independent international and local institutions such as Fichtner/Hytsa/Watertech ⁷ and the State Enterprises Commission (SEC) also concluded that during the management contract period, the level of performance in almost all the systems was poor, especially in respect of reduction in non-revenue water. Further, the supposed or expected technical, managerial, financial and organizational superiority of the private consortium over the public sector assumed under a MC can be questioned. The performance described above therefore brings into question the suitability of the MC in the urban water sector. And apart from these VFM concerns, the MC also raises some critical democratic governance issues with regard to the use of market-type tools to reform politically sensitive social services.

Governance and Policy Outcomes of the MC

We now turn to examining the democratic governance effects of the GWCL/AVRL reform efforts. Concerns with questions of democratic governance are relevant to this study because efficiency gains are too often seen as a trade-off to democratic principles, when in fact they should be seen as two sides of the same coin. One cannot therefore “sacrifice” accountability and transparency in the name of greater efficiency and effectiveness, because true effectiveness would include democratic governance concerns as an end in themselves.

The first governance indicator to be examined is stakeholder participation and consultation in the public policymaking process, which is a key component of governance in any democracy (Krane, 2007). A core objective of governments in the reform process is to “become more responsive to citizens’ preferences, choices and expectations” (Abele et al. 1998: 6), with a clear indicator of success in this area being the degree of stakeholder involvement and engagement; accordingly, the ability to engage, involve and consult stakeholders has been identified as an important component of democratic and open government process in the public sector (OECD, 2005). In the reform and restructuring of the GWCL, little or no public consultation and participation took place, resulting in what can be referred to as a democratic deficit in the reform process. This lack of public consultation, participation and engagement was one of the main points of opposition to PSP reforms by civil society, NGOs and other stakeholders in Ghana referred to above (Abugre, Amenga-Etego and Mizane, 2003). This also explains why these stakeholders have been calling for the termination of the contract and therefore created a hostile environment for AVRL to operate in. Even the limited avenues of consultation were mired in confusion and controversy during the implementation stage of the MC. As one interviewee indicated:

The MC said we [AVRL] should consult the GWCL before doing some things. But it does not say what consult means and, more importantly, if we are to wait for approval of GWCL after consulting before acting. So we all don’t know what exactly consult means, and this has caused a lot of confusion. Unfortunately, we don’t have the mandate and have to go through the GWCL, and the process becomes too long, cumbersome and slow.

Thus, since the official channels of consultation were not functioning as expected, the resulting failure to consult and engage external stakeholders can be seen to have contributed to the poor performance of the MC.

A second common governance indicator is accountability, which tends to focus on following organizational directives and on holding managers accountable for results, outputs and outcomes rather than on rules and processes (Romzek and Ingraham, 2000). Accountability for results, outputs and outcomes instead of rules and processes is a major selling point of the NPM and MCs (Ferlie et al. 1996; Hood 1991). Contracts, according to neo-liberal orthodoxy, create private markets and are supposed to liberate managers of public organizations from the shackles of bureaucratic rules and regulations. In the case of AVRL, the principle of accountability means that managers were supposed to achieve their assigned financial, operational and managerial performance targets as demanded by the contract. However, as the review given above of the financial record clearly demonstrates, the management of AVRL failed woefully to achieve these performance targets.

According to agency theory, the power of compensation provides one way of avoiding opportunistic behaviour and shirking on the part of the agent, thereby ensuring accountability by the principals under contractual relationships (Lane, 2005; Wallis et al. 2007). The principal, in this case, the GWCL, could withhold payments to the agent, AVRL, in order to compel it to be accountable to its principal. The most effective financial tool that the GWCL could use for this purpose was the revenue account. Unfortunately, it was outside the control and authority of the GWCL. As one respondent remarked during the interviews:

Under the terms of the contract, the revenue account is controlled by the Operator [AVRL] not the Grantor [GWCL], which is an abnormality because he who controls the revenue account has more authority and power. As such, the Operator has more control and power over the Grantor, and we are powerless in our efforts to hold them accountable for their financial operations, revenue and expenditure and budget decisions. Not only is this arrangement difficult to implement, it is frustrating, and there is nothing we can do about it.

It can thus be gathered from an accountability and even a principal−agent perspective, that the MC was flawed. This is a sad situation because accountability measures in administrative reforms are also meant not only to improve democratic governance but also to increase productivity and efficiency, improve service delivery and guard against the misuse of public resources and power (Thomas, 1998). During the field interviews, most industry players could not understand why the government decided to grant complete monopoly of the revenue account to AVRL. As one interviewer complained, it is like “giving away our patrimony to foreigners”. Above all, this singular act created much animosity between the staff of AVRL and that of the GWCL at a time when both parties were supposed to be collaborating for the benefit of ordinary citizens, who just needed reliable potable water. This animosity negatively affected the performance of AVRL.

A third democratic governance indicator that is relevant here is transparency, which refers to the availability of relevant information in adequate portions and in easily accessible forms to concerned stakeholders so that they can make informed decisions. In the case of the MC, interviews with practitioners at the two organizations and with other stakeholders confirmed that the negotiation was a three-way affair between the government, the WB and the Operator and that even the final contract document was not made available to the public and other stakeholders. This environment of secrecy compelled Bakker (2004) to lament that “even the World Bank project and evaluation reports and the Transaction Advisors Report were not publicly available” (Bakker, 2004: 45) for stakeholders to provide input or even to inform them about what was happening. The process was therefore the classic one of top-down policy imposition managed in an undemocratic manner by an exclusive circle of a few elites and powerful institutions. This lack of transparency was also a major concern to critical stakeholders in their unanimous opposition to PSP in general and to the MC in particular (Abugre et al., 2002). In line with the secrecy surrounding the original negotiations, AVRL in its day-to-day operations does not publish its performance indicators or other relevant information for public review, nor does it have any mechanism in place to inform its stakeholders and invite their input, despite the fact that they are directly affected by its actions. The question naturally arises, what was the reason for this secrecy and lack of transparency and openness in the contract negotiation process and post-reform period? The answer can be gleaned from a complaint by an interviewee:

There was a lot of hype and misinformation fed to the public about what was in the contract. Initially, when the Operator was coming in, it was projected too much to some extent that everybody felt when the Operator comes, water was going to flow everywhere. But that was not the expectations of the contract. That was not what the contract itself requested of the Operator.

Commenting on this secrecy and the deliberate misleading of the public on such critical issues, another interviewee stated:

In Accra [the national capital], for us to get water flowing everywhere, we need to build new treatment plants, which is the duty of the GWCL, not the AVRL, but this information was not made public. So AVRL was not going to come in with a magic wand, use the same old and dilapidated water facilities and still produce 50 million gallons water to enable everyone to get water in Accra.

While it is normal − and therefore, to a certain extent, acceptable − for companies to miss performance targets, deliberately keeping this information from the public contributes in a major way to the public’s loss of confidence in the political system and in governance in general.

With respect to fairness and equity − two other important governance indicators − the general performance of AVRL raises serious governance concerns, especially since it disproportionately affects the poor in society, who hope that the contract will better their lot. Above all, the use of private sector models and contracts to provide social services raises a larger question, namely, whether “the burdens of human social life [should] be considered public or private responsibilities” (Bakker, 2004: 51). In the reform of the water sector, therefore, the wider question is the extent to which full cost recovery and automatic price adjustments, which are key conditions of PSP, address the concerns of quality, affordability and accessibility of water for all, irrespective of economic and class status. The public expectation was that this MC would alleviate and improve access to potable water once it came into effect. An interviewee directly expressed this expectation and the disappointment at its failure to materialize:

Initially, when the whole thing [the MC] was to start, it was politicians who actually used it for political purposes. Ghanaians initially hesitated to the contract, fearing that their property [water] was going to be sold out [privatized] to foreigners. And the politicians said, we are not selling out, we are just giving it to some people to manage. All we need is water. When the people [AVRL] come, the water was going to flow. But that was not it. Even when they came, the water was still not flowing, and these people are angry.

From these comments, it is clear that the plan to promote and defend the MC was based on half-truths and, in some cases, outright dishonesty on the part of the government. Accordingly, issues of equity and fairness were not addressed, since the supply of water did not materially improve.

The final indicator for consideration is strategic leadership. In the case of the GWCL/AVRL, what is required is a leadership that not only is knowledgeable in contract negotiations but also possesses the expertise to enforce contractual obligations to achieve VFM and to address democratic governance concerns. Under the MC, the GWCL was empowered to monitor, regulate and supervise the activities and operations of AVRL, from which it follows that the “making the managers manage” role was assigned to the Board of Directors and management of the GWCL. Unfortunately, the Board and management were found lacking when their regulatory and monitoring skills were needed the most, for, as was seen above, AVRL never attained any of its performance targets for the entire five years of the contract. This failure raises an important question about the role of the GWCL Board in enforcing the terms of the contract and, more generally, about NPM’s central tenet of holding managers accountable for results. The abysmal performance of AVRL therefore stands as a serious indictment of the leadership roles and capacity of the GWCL Board. As one respondent noted:

The responsibilities were shared, but they [GWCL] have additional responsibilities to supervise us [AVRL] as well. And it takes more time and expertise to supervise us than work on their roles, but the leadership of the GWCL is weak and can’t do its work. That is my fear.

These fears proved to be legitimate, and they rightly question the leadership quality of the GWCL and its capacity to simultaneously “let and make the managers manage”. Larbi (2005) also refers to this leadership capacity challenge in Ghana’s SOE sector by arguing that, in most cases, attempts are not made to improve the quality of management in the SOE reform process and in the adoption of NPM reform tools. Similarly, there were questions about the high turnover of leadership at AVRL, which denied their workers any coherent vision and effective leadership plan. As one interviewee stated:

You realize that within the five years that they [AVRL] came, they changed leadership four times. That wasn’t good. The turnover was too high because when they get complaints, they will change and bring another one.

These democratic governance indicators have a real impact on the performance of AVRL. For instance, the lack of accountability, transparency, and consultation during the contract negotiation process was cited by the NCAP in its opposition to the PSP in the first place. Also, during the interviews, most practitioners stated that they were not involved in the contract negotiations and saw the reforms as impositions. When this happens, organizational hostility to reforms emerges and grows. Also, the issue of the revenue account also emerged during the interviews and created animosity between AVRL and GWCL. Thus, instead of them working together, they were rather undermining each other and engaged in a blame game and finger pointing. Ultimately these factors, collectively and individually, negatively impacted the performance of AVRL.

Taken together, the indicators for both the governance performance and the VFM raise critical questions about the suitability of the MC in the first place and, more importantly, whether something could be done to improve the situation. The next section examines these issues.

Neither Public Nor Private: Context, Task Specificity and Creative Adaptation in Reforms

In this section, the MC framework developed above will be used as the basis for analyzing the problems that engulfed the water sector in Ghana, and hence the failure of MC with AVRL. The May 12^th, 2011, edition of The Daily Graphic, Ghana’s leading daily newspaper and government mouthpiece, carried a banner story that quoted the Minister of Water as saying that the government would not renew or extend the contract with AVRL. The official reasons given for the non-renewal of the contract were that AVRL had failed to revive the GWCL or improve its operations, and that non-revenue water under AVRL’s tenure had shot up to more than 50%, against a target of 26%. In other words, the contract was not renewed due to non-performance on the part of AVRL and its inability to achieve the performance targets established under the contract. As a result, AVRL officially handed over the urban water management operational responsibility back to GWCL on June 6^th, 2011.

Yet − contrary to the non-performance claims of the government − AVRL argued that it had in fact worked with the GWCL to jointly improve the reliability, affordability and sustainability of water services. As an interviewee stated, “AVRL was able to improve the internal operations and management of the sector through computer literacy which was very low before our coming. We also created an Information and Communications Technology (ICT) Directorate to computerise outmoded and manual operational systems”. Thus AVRL seriously disputes the poor performance label of the government. What is clear is that in these competing theatres of defensive claims and counterclaims, the MC experiment in Ghana’s water sector − and, with it, the broader PSP approach − ended in failure.

This outcome raises two fundamental questions: what accounts for this abysmal performance of the MC and, more importantly, what conditions are necessary to prevent similar events in the future? The first lesson to be learnt is the significance of context in the use of market-type reform tools, such as the MC, in administrative reforms. Context, as used here refers to the “existing terrain − the topography over which reforms must travel” (Pollitt and Bouckaert, 2012: 47) and which leave significant imprints that have a shaping effect on administrative reforms. Similarly, context can be seen to refer to the “important preconditions for successfully implementing the new public management approach… [which] should not be ignored by countries striving to correct decades of mismanagement” (Schick, 1998: 2). The totality of the impact of context is that the political, structural, institutional and cultural contexts of any country might therefore enhance or constrain both the choice of reform tool and the final effect of the reform.

From a contextual perspective, a reason for the poor performance in the case of managing Ghana’s water supply can also be found in the political/policy control of public agencies and their inability to devolve power even when they are required to do so under contractual obligations. For instance, the Halcrow and Louis Berger SA reports that gave birth to PSP both stressed that the AVRL should be given enough financial and water pricing autonomy to automatically adjust the prices of water to reflect prevailing market rates in order to achieve full cost recovery. This is sometimes known as full cost recovery in the utility sector. But this never happened. On the contrary, the AVRL had to rely on another state agency, the Public Utility Regulatory Commission ⁸ (PURC) to determine water prices. Yet, even though the PURC was supposed to fulfil this function as an independent agency, successive governments have used their appointing authority to influence water and other utility tariffs by bowing to public outcry and keeping prices below market rates. In this context of external political interference, it is clearly impossible for contracts and managers to function as they should. This lack of financial and automatic tariff adjustment autonomy remains a major challenge for the utility sector in Ghana. Still, with regard to context, it has been argued that public organizations that have “a high degree of face-to-face interaction with large groups of users, like agencies in the welfare and social policy area” (Verhoest et al. 2010: 267), are difficult to reform using neo-liberal and market-based NPM approaches like MCs. The water sector in any country is a conspicuous example of a public agency with a high degree of direct interactions with the public it serves. Accordingly, it is entirely normal for politicians to campaign on the promise of providing their constituents with potable water at an affordable price. In the particular case of Ghana, the water sector is probably one of the most politically sensitive sectors and, for that reason, one whose reform is not best guided by the fixed and unbending terms of a contract. For all these contextual reasons, MCs are unlikely to be, by themselves, the best means to improve efficiency in Ghana’s water sector.

Another factor worth mentioning is the significance of task specificity in the adoption and implementation of MCs. Task specificity refers to the factors that must be considered when choosing a reform tool for a particular organization or policy sector. It therefore raises the practical questions of the suitability and adaptability of individual reform tools, and it focuses on the functional features of particular reform tools or models and their impact on policy outcomes. From this perspective, the consideration of task specificity can be seen to reflect the view that political organization and specific tools of reform are not neutral arbiters in the process of policy adoption, implementation and evaluation, but rather play a decisive role in determining policy outcomes, successes and failures (Hall and Taylor, 1996). In this view, and with regard to the present case, it follows that in order for reform to be successful, the choice of tools must be adapted to suit the specific tasks and informal procedures, routines, norms and conventions embedded in the urban water sector (Hall and Taylor, 1996). Only when this is done, and reforms are crafted to fit the specific tasks and norms of organizations, can positive outcomes be expected.

Task specificity therefore means that not all public agencies − for reasons of their structure, the goods or services they produce, or their relevance and sensitivity to political, social and cultural environments − are suited for hands-off and arm’s-length market-based reform prescriptions (Erridge, 2009). Furthermore, Larbi (2001) explains that the water sector in Ghana “is a politically sensitive public service, making it difficult for the government to allow managers to have full control over operational issues such as tariff setting” (Larbi, 2001: 322). No feature of contractualism can adequately address this basic political issue, because the problem is not simply the straightforward one of management but also has complex political dimensions. As a result, any reform effort that does not take the task specificity needs of public organizations into account is bound to fail. In the present case, the task of the GWCL − water management − is too sensitive to be left entirely in private hands and governed by contracts. Governments therefore need to be extremely careful “when formulating performance criteria or when having expectations about agency performance in measurable quantities, in order to avoid side effects and goal displacement” (Larbi, 2001: 322), which have a demoralizing effect on policymakers and citizens at large.

The final point to be made is the need for policymakers and practitioners to creatively adapt reforms to fit particular organizational needs, cultures and practices. Creative adaptation goes beyond the simple choice of reform tools to consider how best to implement individual reform tools in order to achieve maximum efficiency and other desirable policy outcomes. It therefore entails the successful translation of the aims of policymakers into realistic, actionable plans and their successful outcomes for the benefit of all. Creative adaptation is agency-specific and requires that reformers consider the fundamental conditions and circumstances − such as innovative ideas, appropriate technology, skills and expertise, adequate remuneration and worker motivation − within a particular policy sector or organization. It can also involve other critical policy resources, such as time, funding and even political support.

For reforms to be creatively adapted to fit particular organizational needs, they must be applied through bottom-up approaches, not only by top-down impositions on administrators. With regard to the MC under discussion, it can be seen to have been applied as a top-down imposition because most practitioners were not adequately trained to understand and implement the reform policies. Training, capacity building and continuous learning have all been recognized by scholars as critical both to the reform process itself and to building an improved public service in most developing countries (Schick, 1998). Similarly, other scholars argue that engaging workers, consulting them and giving them a voice in the policymaking process plays a large role in cultivating critically needed institutional support for reforms (Caiden, 1999). However, in the case of Ghana, interviews with GWCL senior management show them to have been uninvolved in the negotiations and development of the MC. Further, and apart from their non-involvement, they were also not taught what new attitudes, demands and changes were expected of them once the reforms were adopted and implemented. The MC was therefore imposed on them with little or no input from management’s perspectives and experiences (Larbi, 1998). In such cases, new reforms typically create a hostile environment and resistance among practitioners because of fears that these new reforms and changes will lead to the loss of jobs. A more detailed and comprehensive training and capacity building program, one that explains the new expectations, demands and cultural changes that will accompany reforms, will go a long way to alleviating the fears of these mid- to low-level practitioners and, in the process, create a more hospitable environment for reforms.

Summary and Conclusion

This article began by asking two important questions about the use of contractualism as a means of reform in the water sector: (1) is contractualism suitable for the reform and restructuring of urban water sector reform activities in developing countries; and (2) under what conditions can contracting actually promote efficiency and effectiveness in the water sector? It reviewed the literature and theory on contracting in the public sector and based on this designed a framework for analyzing MCs in the urban water sector in Ghana. It then examined the VFM and governance performance of the MC involving the government of Ghana and AVRL. The evaluation indicated that AVRL failed to meet its performance targets, and this failure was cited by the government for not renewing the contract upon its expiration. The article then examined some of the local conditions and factors, such as context, task specificity and creative adaptation, which contributed in various ways to the poor performance of the MC.

The empirical and normative evidence presented here clearly demonstrates that contractualism did not prove to be a suitable approach to the reform of water sector and services provision. The study also showed that in order for contractualism to achieve efficiency and effectiveness, certain vital conditions must first be met. The major finding is that context, agency task specificity and creative adaptation are critical factors for success in the use of private sector models for public agency reforms in general and water sector reforms in particular in developing countries. Not only are these elements critical in themselves, together, they represent the most important structural, institutional, political and administrative contingency factors to consider if policymakers are to be successful in the reform process. They therefore constitute the core tenets and undergirding principles for the creation of citizen-centred, democratic, accountable and participatory public organizations that fit the needs of different politico−administrative cultures and systems.

These considerations are important because administrative reform efforts are not carried out in isolation or limited to developing countries. They are mostly driven by the deep-seated core beliefs, values and morals exhibited by policymakers in their efforts to improve the management of public organizations and the provision of public services. Also, reform tools such as MCs are often implemented in imperfect conditions and institutions that have serious managerial, financial and technical challenges. Further, MCs are not neutral reform tools, but ideologically value-laden and, as such, come with some predetermined and ideological expectation of the adopting country or policy sector. All these considerations can potentially create a very inhospitable environment for top-down market-type reforms. Context, task specificity and creative adaptation are therefore critical to successful reform efforts because they address core and critical political, social, cultural, and economic issues responsible for building hospitable environments for externally imposed market-type reforms.

In conclusion, it was ordinary citizens who were adversely affected by the failures enumerated above. Already, statistics indicate that the majority of the urban poor in particular do not have access to affordable and dependable potable water (IFFM, 2002). As one interviewee indicated above, these people were hopeful that at least they would finally have access to water. But their hopes were dashed after all the money, time and other critical resources were invested in the MC experiment. When this happens, it shakes the public trust in policymakers and the policy process as a whole. This adverse effect of the water reforms should therefore be seen as a wakeup call for future reformers to always look before they leap in order to avoid the mistakes of the current reform efforts. If that happens, then this study would have achieved its broader and ultimate objective.