CABRI recently launched a new area of work: Enhancing Digital PFM in Africa, with support from the Bill and Melinda Gates Foundation. The introductory webinar saw participation from 15 African countries, and heard experiences from Rwanda, Benin and South Africa’s experience digitalising their PFM systems. The discussion highlighted key challenges faced by African countries attempting to strengthen these systems: managing, maintaining or retiring legacy systems; overspecification of system requirements; ill-suited and inflexible procurement policies; and stakeholder resistance. These lessons will guide CABRI’s new workstream on Enhancing Digital PFM.
Across the world, countries are looking to improve their government systems and services through digital technologies. One thing that holds true in every setting: digital transformation is complex. It simultaneously requires careful preparation and planning, as well as a change in behaviours and an ability to adapt to unexpected obstacles.
Ministries of finance play a central role in digitalising government. Financial systems lie at the heart of government functions – ensuring domestic resources are effectively raised, allocated and used to meet policy objectives. To effectively link financing and policy objectives, ministries of finance must think beyond merely upgrading Financial Management Information Systems (FMIS). Instead, they need to consider how to make relevant data available when required, and choose the digital solution that allows the functions of PFM to improve. As more digital solutions become available – such as Business Intelligence, Big Data, Artificial Intelligence and Robotic Process Automation – the question becomes how governments can leverage this expanding digital toolkit to better link financial systems with the delivery of goods and services for its people.
This year, CABRI initiated a new programme area called Enhancing Digital PFM in Africa. Across Africa, governments spend substantial time and money implementing new digital systems. This area of work seeks to increase knowledge sharing about these experiences to ensure member countries can more expediently and successfully implement solutions that help in their budget reform initiatives and strengthen the delivery of public services.
On April 18, 2024, 60 participants from over 15 countries joined a CABRI webinar to launch this new programme area. The webinar provided an opportunity for CABRI to start facilitating discussions on Digital PFM, with three countries sharing their experiences in a panel discussion:
| South Africa | Rwanda | Benin |
|---|---|---|
| HENDRIK SWANEPOEL | PLACIDE MUKWENDE | FRANÇOIS BEKPON |
| South Africa’s iFMIS reform was initiated in the early 2000s, but faced substantial challenges that stalled the implementation of the new system. Procurement processes and defining specifications have both meant that South Africa continues to utilise existing Excel-based solutions for PFM processes. | Rwanda shared their experience going ‘paperless’ through their iFMIS. The success of Rwanda’s iFMIS is, in part, due to the in-house team of developers, who are able to adjust the systems as the needs evolve, and deep commitment to digitalisation at senior levels of government. | Benin’s Ministry of Finance explained the value of using a modular approach to building their digital PFM systems – where different software solutions could be used in different parts of the PFM system. The focus for Benin was ensuring a single source of data and all systems followed an established set of guiding principles. |
Over the course of the 1.5-hour webinar, many common challenges were raised and discussed. A few points that stood out included:
Managing, maintaining or retiring legacy systems
Every country has several existing digital systems – whether this is a Commercial Off-the-Shelf (COTS) FMIS or a set of tools including Excel workbooks that enable the production of budgets, reports and analyses. In this modern context, countries are no longer starting from scratch, but must design digital reforms with the legacy systems and processes in mind improving the system module by module.
The South Africa example provides a word of caution about neglecting existing legacy systems. A large scale FMIS reform takes on average seven years to complete in a sub-Saharan African context[1] . In South Africa, the initial reform was conceived in early 2000. With the promise of a new integrated system, departments were dissuaded from investing in their existing legacy systems. This meant that the original Excel-based tools used in the majority of PFM functions have not been reviewed or upgraded. While these simple tools have been reliable and functional to date, Hendrik Swanepoel noted that cracks were beginning to appear due to these legacy systems being neglected.
In Benin, existing systems were maintained as the new PFM systems were developed. François Bekpon highlighted that legacy systems should not be discontinued too quickly. During the digital reform in Benin, there would be periods of time where the new and old systems would run in parallel until confidence in the new systems’ ability to collect quality data and capability to run those new systems was established. Only after the new system was tried, tested and adjusted were old systems retired.
Defining the specifications – the language problem
There is a wealth of research that highlights the need to be adaptive and iterative in the design and development of digital systems[2]. This entails a movement away from linear planning, in which specifications are defined up front and the ability to adapt to users’ experiences and on-going changes in PFM processes can be restricted. It is not always possible to define the exact specifications up-front, but digital systems also can’t be developed without some parameters in place. Countries therefore need a process that continually gathers feedback and input and uses this to adjust the plan, the system specifications and the tools. The objective with this approach is to keep the focus of the digital reform on the ultimate goals it is trying to achieve. This requires strong capacity for both coordination and consensus within government.
One reason defining specifications is so hard, is that digital PFM systems is a multi-disciplinary process. It includes technical IT specialists and a plethora of users – from the economists, data specialists, accountants, transactors. Each user has a different understanding of the needs and purpose of the digital system. While developers try to meet these needs, there are no translators between these experts. Too often, there is a misalignment in understanding and the resulting system falls short of expectations of the different users.
In Rwanda, bringing the FMIS development team in-house has helped address this problem. Placide Mukwende explained that there is a dedicated team in the Ministry of Finance responsible for developing and consistently improving FMIS. This means that when a misalignment of expectations or feedback from users is identified, the team will translate it to developers who will be better able to respond, compared to when the government relied on external service providers for instance.
Procurement for digital reforms is challenging
Specifications play an important role in public procurement processes for digital systems. Existing public procurement rules often require ministries of finance to define their needs up front when contracting out the development of the digital systems – these processes are often in place, and enforced by law, to manage potential risks during the contract-management stage and also leverage economies of scale and save on costs. However, these traditional approaches to procurement can limit governments’ capacity to take an adaptative and problem-driven approach and sometimes end up costing government more than anticipated, especially when bottlenecks occur and specification need to be reconsidered. They also often neglect the large investments needed in upgrading infrastructure (e.g. servers, datacentres, network devices, computers), maintenance of systems and on-going trainings.
Given the multi-disciplinary nature of a digital PFM reform, there is also a risk of overspecification. This happens when teams try to ‘get very specific’, when often the desired change is not yet fully known due to unpredictability around PFM practices and change management. Contracted service providers and developers may deliver on the defined specifications, but are not incentivised or equipped to focus on questioning these specifications, solving problems and delivering the outcomes desired through the reforms.
Benin’s modular approach to their digital transformation helped to avoid issues of over-specification. Rather than defining precisely what each component of the digital system would look like, the country outlined guiding principles for data systems in the public sector. It allowed different solutions to emerge and adapt, while maintaining a unified and interoperable system based on simple but clear data sharing rules (common data specifications, shared platforms).
Managing people
A final common concern in implementing a new digital system lies in the management and capacity of people. Success of a digital reform lies not only in designing and developing a new system, but in considering how it will be used. This requires increasing digital “literacy”, as well as trust in digital data. Even a good digital system, that reduces the need for repetitive activities and improves data sharing is useless in a context where people are not trained, willing or confident enough to use it.
In Rwanda, the move to paperless PFM systems was supported by the most senior levels of government. Ministers were some of the first to move towards digital signatures and approvals. However, the change to paperless saw some resistance as individuals lost their travel allowances as they did not have to physically bring documents to Kigali for signatures. Resistance to change should be investigated to understand the root causes and potential solutions.
Understanding the behaviours, incentives and capacity of the users, as well as who may resist a digital reform, is an essential part of change management. Digital transformation is not simply about implementing a digital system, but requires a careful eye to what users need, what systems can be improved, and who may lose out as a result of the process.
CABRI’s Enhancing digital PFM programme
Much of the discussion during the first webinar focused on country experiences and challenges implementing a “core” FMIS to strengthen PFM systems and improve service delivery. It is essential, however, that we begin to look beyond the FMIS in how we understand digital PFM. At CABRI we recognise that the goal is not to implement a digital system for its own sake, but rather to ensure that digital systems enhance how domestic resources are raised, allocated and used to meet policy objectives. This suggests incorporating the needs of ministries, departments and agencies (MDAs) and understanding how they collect and use financial data to improve service delivery and population outcomes.
With support from the Bill and Melinda Gates Foundation, CABRI has initiated a six-month scoping exercise to understand the gap between financing systems and service delivery decisions, and the opportunities that digitalising presents to bridge this gap. This investigation will shape the future of our work in digital PFM. During this six-month period, CABRI will conduct five in-depth reviews of countries’ experiences in digitalising their PFM systems – looking beyond iFMIS to understand how different data systems are brought together to strengthen service delivery.
These reviews, and a second webinar to be conducted in mid-2024, will be used to design the second phase of CABRI’s work. The goal is to improve knowledge sharing for successful digital PFM reforms. This could be done through a digital component of CABRI’s flagship Building Public Finance Capabilities programme, holding a Policy Dialogue to share country experiences in identifying solutions to the above challenges, or creating an online repository that houses the work done to date on the African continent in digitalisation of PFM. We hope to co-design this next step with our member states, to ensure the solutions match the continent’s needs.
If you are interested in participating in the country reviews or have suggestions for areas of focus for this work, please reach out : giselle.hadley@cabri-sbo.org.
The full webinar recording is available here, along with the PowerPoint presentation used.
[1] Dener, C., Watkins, J. and Dorotinsky. W. (2011). Financial Management Information Systems: 25 Years of World Bank Experience of What Works and What Doesn’t. Washington, DC: World Bank
[2] Including : Long, C., Cangiano, M., Middleton, E., et al. (2023) Digital public financial management: An emerging paradigm. ODI Working Paper. London: ODI
And : Uña, G., Allen, R. and Botton, N. IMF Technical Notes and Manuals (2019) How to design a financial management information system: a modular approach. Washington DC.