The paper essentially argues that aid-dependent countries have uncritically accepted the public financial management reform prescriptions of international aid organizations. Many of the prescriptions are 'new public management' related reforms that might not be appropriate for developing countries, or meet the actual country needs for better PFM. The paper goes on top review two particular reforms --- MTEF and IFMIS -- for Tanzania, Uganda, and Ghana. The details of the cases are worth reading, but they key points from the paper on the reforms are worth reprising.
MTEFs
MTEF's have the potential for linking policy with budgets, and enabling better decision-making. Reviews to date of MTEFs have shown few successes, and the article notes that some of this limited progress is attributed to implementation flaws rather than design or conceptual flaws. Four specific points made with respect to flawed implementation are worth highlighting:
- no clear strategy for moving the entire PFM system towards a performance focus
- unrealistic sequencing or timing of the reforms
- donor or consultant driven reforms (versus government-led), and lack of independent advice to government on the reforms and sequencing
- adding over-specified, overly sophisticated IT solutions that don't match the environment
As an aside, the paper does make a very interesting observation. The paper notes that the United Kingdom has been trying to introduce medium-term budgeting since the Plowden Commission of 1961, and despite repeated efforts over decades, it was not until 1998 with the adoption of the biennial Comprehensive Spending Reviews that a workable approach was found. Thirty-seven years in a high-capacity environment to have a key feature of an MTEF in place should give pause to all those advancing MTEFs for developing countries, and also to government officials about the long-term nature of the commitment to reform.
From the paper, the MTEF country case study lessons are
- For Uganda, budget formulation reforms started early, focused on practical steps to improve budget management in line with the fiscal challenges faced by the country. While hard to assign causality, the improvements in budgeting did along with improvements in macrofiscal stability. Only later, after some years of budget reform, did external parties visit Uganda and inform the authorities they had an "MTEF." However, according to the paper, there is no evidence in Uganda that the MTEF aided strategic reallocation or priority setting -- all sectors benefited from improved macrofiscal stability and economic growth, though education did benefit slightly more.
- For Tanzania, a relatively successful technical/process MTEF implementation built upon (a) considerable external technical assistance (though the paper argues the authorities could have done it without external help) and (b) five years of experience with a Rolling Plan and Forward Budget introduced in 1992. However, in terms of results, the MTEF did not help achieve microfiscal stability (Tanzania had already attained this when the MTEF was introduced), nor has it aided in strategic reallocation of resources across sectors to any appreciable degree.
- For Ghana, the paper argues it was introduced too quickly, based on very limited experience in other countries. Moreover, it seemed to be an effort to directly import the specific approach and techniques from another country into Ghana -- lacking any customization to the Ghanaian circumstances. The paper suggests that MTEF was the wrong focus for PFM reform in Ghana, as basic financial controls and annual budgeting were not working well. Loss of public sector staff expertise to the private sector is also identified as a contributing factor to limited MTEF progress.
IFMIS
The full benefits provided by information technology in developing countries for public financial management have yet to be realized. The paper notes that on the one hand, some believe IT provides opportunities to leapfrog intermediate stages of development. On the other hand, IT systems require higher-skilled human capital to operate, and developing an retaining such skilled human capital has proven difficult. As the paper notes, the success rate of IFMIS systems leaves much to be desired, and could reasonably serve as the basis for challenging some of the advanced IT solutions often adopted.
Another aside for the paper is worth highlighting. The U.K. does not have an 'integrated' FMIS -- each local government, ministry and public sector organization has its own FMIS, using different software and platforms. And, not unlike other countries, the U.K. has struggled with implementing large and complex systems, even within one ministry.
Case study lessons:
- Tanzania introduced IFMIS step-by-step, in a modular fashion, rather than in one go. Automation was targeted to PFM areas considered critical for improving financial management. Some of the teething problems resulted from insufficient attention to the 'soft-systems' surrounding the technology (e.g. linkages to budgeting, commitment controls). The disadvantage of the modular approach has been incomplete data coverage, even as the timeliness of financial reporting has improved. Commitment controls have not been integrated (as of 2005) into the system, and arrears still occurred. A very interesting point in the paper is the absence of capacity in line ministries to use the data --- and perhaps a system designed with more data than the authorities needed --- where the FMIS can produce detailed reports of costs by activity and program, but no one interested or able to use the reports to manage or improve operations.
- Uganda undertook to implement an IFMIS, building on a number of legacy automation efforts of prior years. As of 2005, the system was being piloted, and was still on track, despite the scope of the effort (all central and local governments).
- Ghana started working on an IFMIS in 1996, and some ten years on, has limited functionality to show. According to the paper, a primary reason for the problems was an overly complex initial design. Other issues, such as ownership, human and institutional capacity to manage the project and implement, also contributed to the delays. The paper also suggests that the IFMIS was trying to automate formal rules that were markedly different from the informal practices dominant in the public finance system, and this 'gap' was an indication that progress would in fact be slow in implementing the reform. The paper notes that new project management and governance structures were being put in place in 2004-5.
Conclusions
The main conclusions of the paper include:
- highly aid-dependent countries are at a disadvantage with respect to international financial organizations, and may accept without question some of the reforms being proposed, even if evidence for the reforms effectiveness is lacking.
- fads and fashion in PFM reforms spread rapidly, with 'new' reforms just started in developed countries suddenly being exported globally.
- more country leadership and ownership is essential for reform success.
For MTEFs, Wynne does not draw hard conclusions, only noting the evidence of effectiveness is sparse. For IFMIS, he quotes two pieces of World Bank advice: (i) "Don't fall in love with IT" and (ii) "Think small."
The more general conclusion of the paper is to reiterate the mantra of "getting the basics right" -- thinking small, with pragmatic solutions, and avoiding the cutting edge reforms. A final point of the paper is on the importance of skilled, motivated civil servants for any reform to be successful and sustainable.
[The paper also includes a useful Annex, not addressed in detail here, but which we commend to the reader. It is entitled "Acquiring An Integrated Financial Management System," and is a reprint of an article by Michael Parry that was originally published in Accounting & Business magazine, December 2004. Still highly relevant.]
PFM Blog Comment
Two brief remarks:
- For years the mantra of 'getting the basics right' has been repeated, but, there is surprisingly little on specifically what those basics are. The PEFA performance measurement framework actually starts to identify these more specifically, and perhaps a future piece of work might usefully pull out the 'basics' and present them in that format.
- The papers' main arguments and findings are very consistent with the lessons of PFM reform that went into the Strengthened Approach to PFM reform" and the PEFA Framework (a) country leadership in reform planning and implementation; (b) donor coordination in support of the government reform program; and, (c) a common framework for assessing PFM system performance. On the latter issue, Wynne's paper does not argue for this directly, but the quotes included in the paper, often from the same institution, but with fairly different conclusions about one country, suggest the absence of a common framework is a real problem --- not just for assessing progress, but for just learning what does and does not work. (See our April 21, 2008, blog for more PEFA details.)
==============================================
* The African Capacity Building Foundation (ACBF) -- created in 1991 and located in Harrare, Zimbabwe -- is a collaborative effort of some 32 African and non-African countries and organizations, including the IMF, World Bank, and African Development Bank. The ACBF is dedicated to building institutional capacity in Africa as an essential element to development. And capacity building, of course, is a central issue for PFM system development.
In trying to learn what public financial management (PFM) reforms work, it is important to challenge conventional wisdom and look afresh at what has been tried, and its relative success. Model reforms need to be compared with actual experience, and lessons learned based on evidence need to be drawn and fed back into the reform processes. A 2005 working paper "Public Financial Management Reforms in Developing Countries: Lessons from Ghana, Tanzania and Uganda" by Andy Wynne does just that, looking at medium-term expenditure framework (MTEF) and integrated financial management information system (IFMIS) reforms in the three African countries cited. The working paper was published in December 2005 by The African Capacity Building Foundation.* 
