The integrity management toolbox in action: a study of 22 urban water service cases

The Global Corruption Report 2008, which highlighted pervasive corruption throughout the water sector, remains relevant today, impeding progress in realizing the human rights to water and sanitation (Transparency International 2008). Extensive research consistently reveals a spectrum of corrupt activities within the water sector, including high-level government bribery and corruption (Davis 2004; Plummer 2008; Gonzalez de Asis et al. 2009; Breen & Gillanders 2022). Water utilities are prone to experiencing corruption within the service delivery process (Trapnell et al. 2017). Contributing to utilities vulnerability are the fragmented governance structures, ambiguous institutional frameworks, and role overlap within the water sector (Tropp et al. 2017). The monopolistic nature of water utilities and complex infrastructure create opportunities for corruption in construction and procurement (Butterworth & De La Harpe 2009). Integrity failures within water utilities can manifest as irregular contracting, illegal connections, collusion between officials and contractors, falsification of invoices, and even staff soliciting bribes or sexual favours (Plummer 2008; UNDP-SIWI Water Governance Facility 2017). Compromised top management can lead to biased decision-making, nepotism, demoralized staff, and a lack of transparency (Water Integrity Network 2021).

The costs and consequences of integrity failures vary significantly. Some malpractices impose substantial costs on utilities and consumers due to their prevalence. For instance, a 2020 Afrobarometer survey in 34 countries found that 1 in 5 respondents had paid bribes or performed favours to obtain utility services (Howard & Han 2020). More sophisticated forms of integrity malpractices, such as financial statement fraud or collusion, occur less frequently due to the expertise and coordination required. However, even isolated instances of high-level corruption can result in significant damage. A study analysing corruption red flags in water and sanitation contracts in Latin American countries revealed that projects with elevated procurement corruption risks can be up to 16.4% more expensive in terms of unit costs (Adam et al. 2020). Indirect consequences, including public health effects, environmental harm, productivity losses, and impacts on marginalized groups, frequently exceed the direct financial costs associated with integrity failures (Abu et al. 2021). In other words, the broader and more far-reaching impacts of integrity lapses can be even more significant than the immediate, tangible expenses. Ultimately, integrity failures in water service delivery obstruct progress toward the human rights to water and sanitation (Ali et al. 2023).

Anti-corruption policies in the water sector, like those in other domains, have traditionally framed corruption as a principal-agent problem (Huppert & Wolff 2002; Krolikowski 2014; Bellaubi & Pahl-Wostl 2017). However, there is a growing inclination towards embracing Olson's (2012) collective action theory (from 1965) in the anti-corruption discourse. While the principal-agent framework prioritizes monitoring and accountability from the principal to the agent as a solution to corruption (Klitgaard 1988), it may overlook the leaders' commitment to enforcing reforms (Persson et al. 2013; Bauhr 2016). Conversely, collective action theorists view corruption as a collective action problem wherein individuals, driven by self-interest, hesitate to cooperate even when it is collectively advantageous due to concerns like free riding (Marquette & Peiffer 2015). This perspective highlights the importance of fostering collaboration and trust among stakeholders to establish a shared framework for addressing corruption challenges (UNODC 2013).

Within this context, a range of collective action solutions manifests across various forms and levels in the water sector. At the sectoral level, these encompass anti-corruption policies and strategies often orchestrated by national governments, setting the foundational framework for addressing corruption while unequivocally censuring any instances of free-rider behaviour (Butterworth & Potter 2014). Furthermore, collaborative solutions, exemplified by Integrity Pacts, entail multi-stakeholder engagements (Gonzalez de Asis et al. 2009). Regulators, occupying a pivotal role in promoting integrity, can lay down definitive rules and standards for the water sector's success (Berg 2013). Additionally, utilities frequently engage in certification processes such as adhering to ISO norms or AquaRating (Avello et al. 2019).

The Integrity Management (IM) Toolbox (WIN 2023) is a distinctive framework that blends elements from the principal-agent theory, primarily focusing on bolstering transparency and accountability measures to enhance oversight and control dynamics between principals and agents. Concurrently, it is structured to foster collaboration and trust among staff members within water utilities. To attain this goal, the process incorporates essential collective action variables as identified by Ostrom (2010). These variables encompass the formation of a small and diverse participant group, facilitation of face-to-face communication, and the cultivation of trust-building activities among utility personnel, to counteract any tendencies towards free-riding behaviour. This paper investigates its application in 22 water utilities from 2013 to 2018, aiming to provide evidence on identified integrity risks and selected instruments while assessing the IM Toolbox effectiveness and potential enhancements.

The primary objective of the IM Toolbox is to empower water utilities in effectively addressing the intricate challenges associated with integrity by fostering collective action among utility staff members. This toolbox serves as a comprehensive resource, offering practical guidance and tools specifically tailored for water utility managers.

The IM Toolbox operates through a well-structured three-phase change process, characterized by a concise timeframe for efficiently assessing and managing integrity risks. In the initial preparatory phase, the toolbox is customized to suit the unique context of the utility. Furthermore, this phase emphasizes the meticulous selection of workshop participants and key stakeholders. The approach here ensures diversity within the group while limiting the participant count to a maximum of 40. Simultaneously, the planning phase actively engages participants in a workshop for a comprehensive analysis of current risks, the choice of relevant tools to address the most pressing risks, and the development of a detailed roadmap for their application. Subsequently, the implementation and monitoring phase takes effect, involving the allocation of responsibilities, the roll out of the agreed action plan and the documentation of progress. The seven steps encompassed by the preparatory, planning, and implementation phases of the IM methodology are described in Table 1 (Hermann-Friede et al. 2014):

This study adopts a structured framework aligned with the IM Toolbox steps detailed in Table 1, enabling a comprehensive analysis of six dimensions (Table 2). The first dimension assesses Identified risks examining integrity risks unique to each water utility. Integrity risks refer to factors that hinder organizations from fulfilling their mandates with integrity and compromising sustainable water service provision. Secondly, Integrity tools encompass measures and actions taken by each water utility to mitigate or prevent integrity risks, transforming them into opportunities for positive action. The level of implementation for each integrity instrument is classified into four categories: (i) fully implemented, (ii) partially implemented, (iii) not implemented, and (iv) no information available. Implementation levels were assessed 12 months after the workshop. Products are defined as the various policies, programs, guidelines, tools, capacities, and systems selected as risk mitigation tools and included in the roadmap (Step 6 of the IM Toolbox methodology). Impacts investigates medium-term effects and overall utility improvements resulting from tool implementation, including enhanced governance and service delivery.

The analysis relied on four main data sources, as shown in Table 2:

• Primary sources: These included comprehensive reports generated during the IM Toolbox's application across various utilities. These reports encompassed the methodology, roadmap, and follow-up reviews conducted by appointed coaches.

• Secondary report: A supplementary source involved a secondary report that scrutinized the application of the IM Toolbox within three Latin American utilities (Barreto-Dillon et al. 2018).

• Inputs from international coaches actively engaged in IM processes. The contributions of eight of these facilitators enriched the data collection and the discussion.

• Interviews with utility managers: To evaluate the impact of the IM Toolbox's implementation, interviews were conducted with utility managers. These interviews were further supported by publicly accessible information concerning the integrity progress of the utilities. However, it is important to note that, due to personnel changes, certain managers were unable to provide insights regarding long-term impacts. As an ethical practice, all interviewees were fully informed about the research's objectives, and the confidentiality of responses was assured. The scope of the interviews exclusively pertained to the application of the IM Toolbox.

The study acknowledges two limitations. Firstly, the availability of long-term evaluations and changes in key staff hindered comprehensive long-term follow-up. Secondly, the lack of standardized approaches for data collection on IM Toolbox implementation and varying report quality posed challenges. In addition, due to the sensitive nature of the information and to maintain confidentiality, the individual risks and instruments are not attributed to the utilities with their names, and the evidence of long-term impact has been anonymized. Consequently, considering the limited number of utilities per country, addressing discrepancies between countries was not pursued to safeguard the anonymity and confidentiality of the analysis. Instead, the focus remains on presenting aggregated findings and trends applicable across multiple utilities or countries collectively.

The study encompassed 22 utilities across 7 countries on 4 continents (see Supplementary Material: Annex 1). The utilities were distributed as follows: Albania (five utilities), Kenya (seven utilities), Laos (four utilities), Bangladesh (two utilities), Cambodia (two utilities), Ecuador (one utility), and El Salvador (one utility). These utilities were selected based on the availability of information and the ability to contact personnel for data collection purposes.

The implementation of the IM Toolbox began in Kenya in 2013 and expanded to other countries subsequently. Among the utilities, EPMAPS in Quito, Ecuador, and ANDA in San Salvador, El Salvador, were the largest in terms of household connections and employees and can be considered large utilities, with over 500,000 household connections. WSU Berat-Kucove and Lushnje in Albania, Mombasa (Kenya), and Chittagong (Bangladesh) can be considered as medium, with a number of household connections between 75,000 and 105,000. The 16 utilities left can be considered as small utilities, having between 4,000 and 35,000 connections each.

In this section, we present the study's results, encompassing risk identification, instrument selection, and implementation progress using the IM Toolbox. The risks and tools are categorized according to different areas. Governance and Management, encompassing policies, rules, practices, and processes governing an organization, including ethical standards, regulatory compliance, and accountability; Human Resources, addressing personnel management aspects like staffing, recruitment, training, career development, compensation, benefits, and staff motivation; Financial Management, involving the strategic oversight of an organization's financial assets; Contract Management, which covers the entire contract lifecycle, including procurement and compliance; Operations, encompassing daily water service provision activities, such as asset management, routine field tasks, infrastructure monitoring, maintenance, and repairs; Consumer Relations, outlining strategies and processes for managing interactions with water service consumers; and External Environment, focusing on external factors like legal frameworks, socio-economic and political conditions, and natural disasters, all of which can impact governance, operations, and service delivery.

The water utilities involved in the study identified a total of 68 integrity risks during the implementation of the IM Toolbox. These risks were categorized by type and frequency, as detailed in Supplementary Material: Annex 2. Meter tampering and falsified meter readings, classified under Operations, emerged as the most prevalent risk (n = 8). Consumer Relations exhibited two risks that were selected six times: negative consumer perception of service providers and inadequate complaints management. External Environment followed, with vandalism, theft, and external interference as the most prevalent risks (n = 5). Aggregating data by area showed that Operations had the highest risk selection frequency (28%), followed by Consumer Relations (27%) and Human Resources (13%).

Utilities selected a total of 90 tools for implementation (see Supplementary Material: Annex 3), averaging 1.3 instruments per identified risk. These tools aimed to address various aspects of integrity issues. In the Human Resources domain, commonly chosen instruments included field inspections and staff supervision (n = 6), a sanctions catalogue for unethical behaviour (n = 4), and integrity training for staff and management (n = 3). Within Consumer Relations, the most frequently selected instruments were consumer complaint management (n = 9) and strengthening consumer feedback (n = 8). In the Operations category, the development of meter reading procedures (n = 5) was prominent. Additionally, utilities placed significant emphasis on increasing citizen participation in five instances. Overall, utilities showed a greater inclination toward Human Resources instruments (29%), followed by Consumer Relations (22%) and Operations (21%).

Following the selection of instruments, a roadmap was finalized, and implementation took place within a period of 6–18 months. This review assessed the implementation progress after 12 months. Among the studied utilities, 64% of the instruments were fully implemented, 9% were partially implemented, 8% were not implemented, and data were unavailable for 19% of the cases. The levels of implementation varied across different areas. While Financial Management instruments were fully implemented in only one-third of the cases, Contract Management instruments achieved full implementation 50% of the time. In contrast, Governance and Management instruments were either fully or partially implemented in all cases. When looking at the implementation across utilities, 50% of them (n = 11) had successfully implemented all selected instruments after 1 year. An additional 18% had fully implemented 70% of the instruments, indicating satisfactory performance in over two-thirds of the surveyed utilities. However, 32% of the utilities (n = 7) reported a lower level of implementation, and one utility reported no implementation of any instrument.

Assessing the transformative changes brought about by the IM Toolbox implementation is complex, especially in the long term. Below are examples illustrating the transformative journeys of selected utilities:

• Utility number 2 initially identified Consumer Relations and Operational issues as significant risks. After 1 year of IM Toolbox implementation, the utility achieved a 100% implementation rate for their roadmap. They improved their complaints mechanism, implemented procedural guidelines for field inspection, and emphasized anti-corruption measures through internal rules and incentives. Their IM project was even recognized with an Integrity Award from the national government.

• Utility number 3 faced procurement delays, bidding problems, weak tariff systems, and an unsatisfactory complaint mechanism. They implemented guidelines for procurement, anti-corruption measures, and codes of conduct. In the span of a year, 70% of the chosen tools were put into action, leading to enhancements in bidding procedures and the complaint management system, including the establishment of response times and policies for addressing customer issues. These enhancements contributed to a slight increase in consumer satisfaction, which rose from 80 to 82%. However, illegal practices continued, and the utility faced an investigation for possible embezzlement 3 years later.

• Utility number 9 addressed issues with connection and billing procedures by implementing key instruments, achieving almost 80% implementation within a year. They enhanced their billing system, consumer tracking, and connection procedures, leading to faster processing times and improved revenue collection. Public hearings engaged consumers in discussions about service improvement, and the utility's process for establishing new connections was significantly streamlined.

• Utility number 20 tackled insufficient complaints management, external interference, and improper reporting. They engaged employees in the transformation, strengthened consumer feedback, reviewed meter reading procedures, and promoted citizen participation. Inter-departmental cooperation improved, resulting in enhanced water supply services, reduced complaints, a 2.3-fold increase in connections over 3 years, and improved infrastructure and debt collection.

Regarding the instruments selected to addressing these risks, Human Resources is the most preferred area, with a higher frequency of selected tools compared to risks. Many tools aim to enhance ethical behaviour aligned with a collective action approach, including staff engagement, role clarification, rotation, integrity training, and codes of conduct (see Annex 3). Governance tools focus on the development of anti-corruption policies and procedures for reporting unethical behaviour. Strengthening Consumer Relations and mitigating risks between staff and consumers, such as improving meter readings, are also emphasized. Financial management tools were both rarely chosen and challenging to implement, potentially due to the time and capacity required for financial changes or blockages at higher levels. Governance tools related to codes of conduct and unethical reporting procedures were nearly fully implemented, although the long-term behavioural impact could not be observed within the investigation's timeframe. External environment tools focusing on citizen participation were also highly implemented.

The success stories and feedback from interviewees underscore the transformative potential of the IM Toolbox in utility management. It served as a catalyst for change, bringing together staff members and motivating them to initiate improvements. The collaborative nature of the process provided a platform for departments to collectively analyse, plan, and address known issues. External support, continuous coaching, and staff training played vital roles in successful cases, enhancing staff understanding of the change process. Top-level management's commitment and involvement proved crucial in initiating and driving the process. For water utilities, political support, leadership, and staff involvement are key to a successful process management change (Martins 2014).

However, not all implementations of the IM Toolbox achieved success. Around one-third of utilities reported low implementation levels. Limited ownership of the process emerged as a key weakness, with some staff members not fully embracing the changes. Communication gaps were identified as a significant issue in certain cases, where the change process was not effectively communicated internally or externally due to capacity limitations in the communication department. The need for an in-depth communication of the vision in change management is one of the key steps of a successful change process (Stouten et al. 2018). Inadequate authority and support for the champions responsible for implementing the instruments, coupled with staff turnover at medium and senior levels, posed challenges to sustaining the change process in the long term. This fact is aligned with evidence from other studies in public services showing the key role of champions in an innovative change process (Shaw et al. 2012; Miech et al. 2018).

The analysis of IM Toolbox implementation presented in this paper offers valuable insights for those involved in IM change processes. The IM Toolbox has proven to be an effective tool for driving collective action improvements among the studied service providers. Within 1 year after the workshop, a significant majority of utilities demonstrated a commendable level of implementation for the integrity instruments.

However, the analysis also highlights the importance of considering additional elements to inform the selection of risks and instruments. To optimize the efficacy of the IM Toolbox, engagement of a wider spectrum of personnel and users in the process of identifying integrity risks is recommended, with the inclusion of anonymous surveys to uncover any potential oversights within the self-selection method. The success of change processes varied greatly, with internal champions, top management support, and ongoing external facilitation and coaching playing crucial roles. In some cases, the implementation of the IM Toolbox sparked a broader cultural shift towards integrity, while in others, staff and management rotations hindered the progress.

To mitigate the influence of these factors, two recommendations are proposed. Firstly, linking the IM Toolbox to other change processes and utility integrity requirements, like ISO norms certification or AquaRating (Avello et al. 2019) can enhance its effectiveness. This approach has been successfully implemented by certain utilities, such as one in Latin America and others in Laos, where the integrity process was integrated with AquaRating certification and ISO norms, respectively. Secondly, regulatory support is crucial to reduce reliance on individual champions and goodwill within the utility. Ongoing efforts with regulators in Latin America to develop mandatory integrity indicators for utilities in those countries are a promising step in this direction. Drawing lessons from these experiences can contribute to fostering integrity improvements in the water sector.

A.J. and P.A. conceptualized the whole article; P.A., A.J., and L.B. developed the methodology; L.B., J.H., L.H., and L.F. rendered support in data curation, P.A. and A.J. rendered support in formal analysis; P.A., A.J., and U.A. administered the project; A.J. supervised the work, P.A. and A.J. visualized the whole article; A.J., P.A., U.A., L.B., and J.H. wrote the original draft; P.A. and A.J. wrote the review and edited the article.

The authors gratefully acknowledge the financial support provided by the Swedish International Development Cooperation Agency (SIDA) to the Stockholm International Water Institute. The funders played no part in the design of the study, data collection and analysis, decision to publish, or preparation of the manuscript. The views expressed in this publication are solely those of the author(s) and do not necessarily reflect the views of the Swedish International Development Cooperation Agency.

The authors express their gratitude to the following individuals and organizations for their valuable contributions: Nagnouma Kone from the Water Integrity Network for assisting in gathering additional data from utilities in Kenya; Avi Sakar and Buahom Sengkhamyong from the United Nations Development Programme for their support in collecting data from utilities in Laos and Cambodia; Ricard Giné and Hélène LeDeunff for their thorough review of the document and guidance; and Christian Fischer from Water Science Policy for editing the first draft of the document. The research team would also like to extend their sincere appreciation to the utility respondents who generously shared their experiences and insights on the topic.

All relevant data are included in the paper or its Supplementary Information.

The Integrity Management Toolbox methodology was developed by Cewas, WIN – Water Integrity Network, and GIZ – Deutsche Gesellschaft für Internationale Zusammenarbeit, with additional contributions from partners such as SIWI. The majority of this paper's authors are (or were) employed by these institutions. Although they played a role in the development of the methodology, they hold no affiliations or financial interests in the subject matter.