« Deux années d’assistance technique en gestion financière publique à l’AFRITAC de l’Ouest | Main | Brazil: Raising the Bar on Fiscal Transparency »

August 18, 2011

Certifying PFM Systems for Donor Budget Support to Fragile States – Professor Collier’s Proposal

Posted by Tej Prakash

In a recent op-ed (later also presented at an Overseas Development Institute (ODI) meeting), Prof. Paul Collier has put forward the argument that where donor aid is allowed to flow through the budget system of a fragile state, it has largely failed to deliver the results promised. The reasons given for this failure, range from incompetence to corruption. And, it is suggested that in the near future, there seems to be little chance of any meaningful improvement in these outcomes. He argues that the governance system in many of these countries is broken, and its focus is by no means primarily to provide services to the citizens. It is suggested that budget systems of these countries are extremely ‘leaky’ (‘looting of the public purse”) and that donors do not have, by and large, either the information or the technical expertise, to prevent misuse of aid money.

Collier makes a distinction between aid given for critical operations and for more general budget support operations. For critical operations he recommends using ‘imported administrative capacity’ to manage all spending, including donor funds through specific project support arrangements. His proposal relates only to donor budget support and does not address existing parallel project based arrangements operated by many donors. He does suggest that it is possible to improve the ‘technical’ aspects of donor flows. The focus of technical improvements would not be to introduce policy ‘conditionalities’ through a back door, but to enforce the country’s own laws. He cites the insight of Tinbergen that to implement any objective, there should be a distinct instrument with its distinct effect. Hence, the two main objectives: meeting the need for funds (how much) and ensuring their effective use (on what), should be managed by two different instruments.

His proposal is that donor aid through the budget should  flow to countries only after the reliability of the budget systems is ‘certified’ to meet a minimum benchmark which can be established by some kind  of standards board. This ‘certification’ can be provided by accounting firms, with oversight and quality control from the IMF and the World Bank.

Some issues with the proposal

Collier’s argument arises out of a sense of donor fatigue in the face of continued governance problems (or, in plain language, corruption) in receiving countries. And the suggestion that continued efforts, almost always donor led to improve domestic systems, have failed in most of these countries over the past decades.

In my view there are some problems with the argumentation.

Collier argues that critical support operations can continue outside the national budget systems and be managed by ‘imported’ manpower. In other words, donors would finance and run projects outside the government, much like many non-government organizations (NGO). However, sovereign nations are different from NGOs. On the political-economy side, it would raise issues of sovereignty and undermine ownership. And on the development side, it will do little to build necessary capacity for future governance in the country. It would also perpetuate the system of dependence. Existing parallel donor executed arrangements, although demonstrating some success in the short run, usually fail in longer term due to lack of sustainability. It is difficult to see how a new batch of externally run projects would be sustainable in a fragile political environment where the basics of governance such as the budgetary systems have crumbled. Hence, the main counter-argument against the Collier proposition is that it is neither feasible nor possible to have two states of affairs inside a sovereign country: one that is fragile and at risk of totally crumbling; and, another a small island of projects run by outsiders.

How are budgetary systems assessed currently?

There are many instruments for assessments of budget (or public financial management (PFM)) systems that donors use. Many donors and multinational organizations, including the IMF, do not use any particular methodology, however, to assess the ‘leaks’ in the system. Moreover, the assessments used do not necessarily aim to plug the ‘leaks’. The broader ‘diagnostic’ assessments generally take a two pronged approach. They analyze the overall PFM framework and determine where the procedural and capacity weaknesses are, and suggest an action plan with suggestions for prioritization and sequencing of reform. Sometimes, especially when the country is participating in a Fund program or requires budget support from another donor such as the Bank, the priorities of the program drive the reform agenda. For example, in the case of Fund programs elimination of arrears of payments is always a high priority item, notwithstanding the fact that payment arrears are symptoms of different problems and liquidating the stock of arrears do not necessarily fully address the underlying problems. Frequently, fragile countries have multiple diagnostic recommendations from many different multilateral and bilateral donors, each with overlapping or even differing sets of recommendation. At the end of the day, the country will make a rational or rather constrained rational choice: this may involve going with the donor which gives most money or where any attached conditions are most likely to be achievable.

Prof. Collier and others mention two instruments: Reports On Standards and Codes, fiscal module (or fiscal ROSC) and PEFA.

The ROSC was not designed primarily to be an assessment of operational weaknesses of budgetary systems. Its main focus is fiscal transparency. It does not lend itself directly to strengthening the operational budgetary processes which can plug the ‘leaks’, except as the effect of greater transparency.

PEFA (Public Expenditure Financial Accountability) assessments are emerging as the broad PFM diagnostic instrument of choice. It is being supported by OECD DAC, and some key multilateral and bilateral donors. There are issues with using PEFA; it also does not drill down enough to address the ‘leakage’ issues. PEFA was designed precisely to provide an assessment of whether a PFM system could meet the accountability requirements of donor governments on the use of their money. This is done by assessing the PFM framework of a country. PEFA assess the PFM framework through rating 28 indicators which cover six broad categories of PFM from budget formulation to budget execution and it includes indicators on procurement and on revenues. It also includes three indicators on donor financing. The assessment points out the relative weakness of various different indicators. These indicators follow a standard methodology, and thus enable some cross-country comparison and a disaggregated comparison and analysis as well as tracking of performance over time.

However, the PEFA assessment does not directly address the problem of ‘looting of the public purse’. The underlying assumption is that if PFM systems are strong, these will prevent misuse or the ‘loot of on public purse’ as Prof. Collier puts it. Is this assumption correct and does empirical evidence support it?

I would argue that this is a mistaken premise. Public sector governance is a complex area, and PFM systems are only one part of the web of public action and operations. Corruption can take place in many ways, and even very robust PFM systems will only raise the bar and the transaction costs associated with in appropriate government action, not prevent it. For example, while there is no empirical study, the establishment of Revenue Authorities in many countries in Sub Saharan Africa with practices operating in line with PEFA/ROSC standards has not eliminated rent seeking activities in these countries. Nor has the percentage of revenue collected as a percentage of GDP risen significantly.

We would argue that enforcement of existing laws, and a contextually robust legal system and the development of institutional capacity, are also required to prevent leakages.

Other areas that impact budgetary outcomes

Reforming PFM alone is not enough. Many factors affect budgetary outcomes, some of which are outside the institutional framework of budgeting and budget institutions including political governance, leadership and management.

Additional factors relate to the competence of the donor community. For example, despite there being donor coordination committees in many countries, there are overlapping and often conflicting mandates and the supply of overlapping or even conflicting technical assistance. Donors competition to provide TA is also well known. Even though action plans are often drawn up where the donors and the country are involved and where responsibility of each donor is well defined, these conflicts and overlaps still happen. Many times, more than one ‘action plan’ exists for different donor priorities.

Donors have different incentives driving them. In cases where a donor is financing projects, there are two important priorities: have an oversight and accountability feed back to the home country of the donor; and, for the project manager pushing a local portfolio is critical to his incentive system. Hence, we see fragile countries with very limited capacity trying to manage a highly sophisticated computerized financial management system at a huge cost. This is also usually undertaken within a framework of externally imposed systems that are contextually inappropriate and which serve to undermine local decision making and management capacity.

 Despite the OECD/DAC efforts and numerous attempts to ensure donor coordination, there is still little evidence of it in most fragile states. In many of these countries, even where donor coordination committees exist, donors work at cross purposes or in competition with each other, resulting often in completely opposing advice. Often the incentive structure of donor assistance is tilted more towards the donors rather than the country. For example, where the incentive for a donor bureaucracy is to build its loan portfolio in a country, it often pushes loans for expensive and unsustainable projects (highly sophisticated computerized financial management projects, for example), knowing full well the country has little capacity to manage it. In the process it diverts huge resources, with tremendous opportunity cost

Hence, before all the blame is laid at the door of the fragile countries or at a particular methodology for the delivery of donor financing, donors, as a group, should look within and deal with their own incentive systems which can be detrimental to the recipient country interests.

The way forward

There are two aspects of PFM reforms driven by the donors. We argue that certification of the budget is the wrong way to go for many reasons. It undermines the legitimate authority of the country’s ministry of finance (MOF), and would weaken the development of capabilities necessary for making budgetary decisions. There is evidence (Prakash and Cabizon, 2008) that stronger MOFs lead to better fiscal outcomes such as lower deficit and debt levels. Collier’s position also contradicts the main message of OECD/ DAC to empower local systems, strengthen ownership and commit to using these systems as much as possible.

PFM is much larger than an accounting issue, even thought accounting is an important part of budget systems (‘what gets measured, gets done’). We argue that the starting point must be for donors to improve aid processes at the donor’s end, including increasing donor funding predictability and avoiding donor competition, overlapping and often overwhelming reform initiatives in counties with extremely limited capacity even for basic tasks; listen and engage more closely with the countries, and above all , taking joint accountability for reforms, and not passing the buck to the countries themselves or evolving parallel systems that compromise capacity development and accountability .

In PEFA an instrument already exists to benchmark the PFM systems and to measure system development over time. Critics of PEFA have suggested that it does not drill down enough into the systems. We suggest that it is an evolving instrument. However, despite noises to the contrary of PEFA being a country led initiative, in actual practice it is often de facto largely donor led. This must change and it can only if the donors understand better a whole dimension of issues, including what works in a particular society and why. We also believe that asking accounting firms to ‘certify’ a budget or the budget process is a wrong approach for two reasons. First, the issue for the reasons outlined above goes far beyond accounting alone, and is far more complex. And secondly, accounting firms do not typically have the expertise or the country view to understand and deal with the public sector in the same way that they might do with the private sector.

Download Op-Ed

Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.


TrackBack URL for this entry:

Listed below are links to weblogs that reference Certifying PFM Systems for Donor Budget Support to Fragile States – Professor Collier’s Proposal:


I am inclined to agree (partially) with the proposal of Prof Collier. Certification will mean the intent to set up and sustain a working, transparent and accountable system. In the transition economies and newer nations, it is important to have a vision of the political executive, backed by the legal authority, for effective and efficient PFM Systems. The integrity and transparency levels and standards are very often compromised in spite of countries assessed by the donors!Donorsshould be having a genuine concerns on those.

After certain levels of maturity (that should be specified for an essential short-term), the national systems should prevsail with the safegurads afains fiduciary risks and modern methods of PFM.That's the time when the excellent views by Tej shall prevail! Amen!

Professor Collier's views are extremely short sighted and retrogressive. If his views were to prevail the development of local systems will never take place. I believe, Tej, you have plainly explained the implications of this knee jerk reaction to beneficiary country corruption in the use of donor funds.

The certification of PFM systems may need further discussion. I would prefer a rating systems using possibly a modified and enhanced PEFA assessment. Such PEFA assessments should be performed in collaboration with officials from the country under review,in order to obtain greater buy-in and ownership of the reform process.

As it was so correctly pointed out by the blog, the fight against corruption runs deeper than the efficacy of the PFM system. Strong leadership at the legislative, policy and technical levels is required to begin to address this issue.

It's always interesting to get an economist viewpoint on public financial management systems. Compelling arguments, but one wonders how far reality is being abstracting from the practical.

Case in point: how much aid is wasted through high transaction costs via multiple organizations rather than the developing country government? How fungible and untraceable are funds once converted to cash in countries under stress? We need a baseline that considers how inefficient and ineffective aid becomes when off budget. Then we need to consider how inefficient and ineffective aid becomes when on budget.

There are a number of "holistic" arguments by Professor Collier. Improving one aspect of public financial management does not necessarily translate to an overall positive outcome. Many factors are involved. There's more to it. Agreed. But, let's think from a practical perspective for a minute. The notion of donors "certifying" a budget sounds altogether dubious. After all, could a donor agree to macroeconomic assumptions? To government objectives? But, what if budget formulation and execution was fully automated, followed international standards and was fully auditable? Corruption occurs in even developed countries. Surely the "efficacy of the PFM system" can provide strong disincentives for corruption - for one thing, corruption becomes much harder to accomplish and much harder to hide with tamper-resistant computerized systems.

There seems to be a meme in PFM that donors are somewhat (possibly primarily) responsible for inflicting poor practices in developing countries. This is described in this entry. For example, many donors force recipient governments to manage funds in separate accounts rather than a Treasury Single Account. This is an underlying problem. And, it seems to support the notion of direct budgetary support as a mechanism for coordination. So, improving aid predictability might not be as effective a strategy as improving coordination. Yet the argument seems to suggest the opposite.

There seems to be a general assumption that capacity is static. Yet, sequencing PFM reform is considered a good practice. To withhold direct budgetary support because of a lack of capacity appears to be a life sentence (or Catch-22). Capacity can increase, and more rapidly than it did in developed countries. But, not if responsibility is withheld. A more practical approach is to sequence budgetary support based on increasing capacity rather than wait until some magic moment when capacity is deemed to be sufficient.

There also seems to be a misunderstanding of the long-term effects of PEFA assessments. Yes, there are differences between country-led and donor-led assessments. Yes, many early assessment were "optimistic". Nevertheless, PEFA assessments have proven to be compelling events that ratchet up positive reform efforts in government and coordinate donors. And, it can coordinate efforts on what is important in the short and medium term.

Dear Tej,

Many thanks for this thought provoking post. I am working quite a lot these days in Haiti and the issues Collier and you raised are especially relevant there. I am nevertheless quite surprised that you or Collier never mentions the fact that almost every country, even fragile, is equipped, at least formally, with “an independent process of scrutiny to certify public spending systems”: it’s called the supreme audit institution. Of course, they might lack independence and capacity in practice but are not the World Bank and to a lesser extent the IMF development institution? Why not, instead of putting together a complex and costly certification mechanism, with the sovereignty issue you rightly mention, invest in building capacity of government external audit bodies in fragile state? If the answer is that this is a lost cause, then we’d better give up budget support, project support and altogether developmental work in fragile states.
Take care,


Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Back to top of page
©2007 IMF. All Rights Reserved. About Us | Terms of Use
/************* DO NOT ALTER ANYTHING BELOW THIS LINE ! **************/ var s_code=s.t();if(s_code)document.write(s_code)//-->