Do ‘good practice’ PFM systems lead to better outcomes in fragile states?

US soldiers and bulldozers clear and improve Route Agha in Afghanistan (Arctic Wolves/Flickr/CC BY 2.0)
US soldiers and bulldozers clear and improve Route Agha in Afghanistan (Arctic Wolves/Flickr/CC BY 2.0)

The questions keeping me awake at night

As I wrote here, there seem to be important unanswered questions as to whether the standard ‘good governance’ reforms recommended by development agencies should be expected to make things better in conflict societies, where the broader institutional context is entirely different.

In my new paper (here), I apply some of this thinking to the ever-glamorous field of public financial management (PFM). Development agencies tend to recommend a highly standardised set of systems and processes when reforming PFM in developing countries, often without much regard to important contextual differences. Countries are advised to adopt budget processes, reporting systems, and accountability arrangements that look like those in place in OECD countries, even when the countries themselves are at radically different levels of development. This is well-established in the broader literature, and I’ve previously explored how standardised PFM solutions are often a poor fit for very small states (see here and here).

Are standard solutions the right ones for fragile states, or are there systemic reasons why they might not be?

Pushing on a string?

I think there are big questions here. Most basically this is because standard PFM solutions rely on broader institutional underpinnings that are – by definition – absent in fragile states. Efforts towards creating PFM systems in fragile states that look more like those in OECD systems are ‘pushing on a string’ because the transmission mechanisms that translate ‘good’ PFM systems to ‘good’ outcomes are missing.

To illustrate, it’s useful to think about what PFM systems do. I think they serve three basic functions. Firstly, PFM systems structure decision-making. Good PFM systems ensure that decisions regarding the use of fiscal resources are made by individuals with the formal mandate to make those decisions, often using appropriate standardised and pre-specified criteria. For example, procurement would be carried out on a competitive basis according to pre-specified selection criteria, or the annual budget would be formulated in accordance with the policy goals of policymakers. Secondly, PFM systems support public transparency. Good PFM systems ensure that adequate information is made available to the public regarding the planned and past use of fiscal resources to allow for public accountability. Finally, PFM systems provide information to officials and policymakers. Good PFM systems ensure that adequate information is available to policymakers and officials to allow well-informed decision-making in the pursuit of public service delivery and fiscal sustainability.

The problem arises when thinking about the transmission mechanisms between these functions of a PFM system, and actual improvements in efficiency and sustainability of public resource use. Good systems will only translate into improved outcomes when a number of conditions are met, and those conditions are very seldom in place in fragile states.

If we assume…

Firstly, PFM reforms are often based on the assumption that new laws, regulations, rules, and procedures will actually be followed. In fragile states, however, such rules are very often ignored. Power is contested between various elites, many of whom have access to the means of violence. A defining characteristic of fragile states is that there is no way for the state to systematically enforce rules over elites or individuals and groups that enjoy elite support. Those with power cannot be compelled to follow rules that are not in their interest, or sanction their supporters for breaking those rules. Having good rules that are not followed is very common but doesn’t really help.

Secondly, providing information on the use of public resources to the general public is most useful when the public uses its voting power to hold policymakers accountable for the delivery of public services. Providing more public information generates incentives for better use of public resources because policymakers know that they will be held accountable for the quality of services and economic management when it comes to the next election. But this is very seldom the case in fragile states, where voting patterns are dominated by patronage (i.e. the distribution of state resources and rent opportunities to powerful individuals in return for their electoral support and that of their followers). The accountability loop is further weakened when – as in many fragile states – many services are provided by aid agencies, and voters lack sufficient education or literacy to meaningfully engage with the public finance information that is being provided.

Finally, providing better information to officials and policymakers through the PFM system is useful when policymakers and officials are systematically encouraged to improve services. But, as above, incentives on policymakers to use available information to improve services are very weak if they are elected on the basis of patronage rather than service delivery. Further, public sector appointments and promotions are often based on political allegiances and patronage, rather than on rewarding those who drive improvement. In this context, while many officials in fragile states are highly motivated to improve services and deliver results, this is often despite, rather than because of, any incentives arising from improved information generated through the PFM system.

What is to be done?

Of course, it’s still important that fragile states have good rules and laws, provide information to the public, and ensure that policymakers know what is going on with public finances.

But we need to think more about the specific causal mechanisms through which PFM reforms should be expected to lead to improvements in the efficiency and effectiveness of public expenditure. Recent work on problem-driven approaches might help inform this kind of thinking. Rather than beginning with an assumption that fragile state PFM systems should be reformed to look like those in OECD countries, it might be more useful to: i) think about the specific constraints that are leading to inefficiencies; and ii) consider a range of options for addressing those constraints, taking careful account of the broader institutional context and what this might mean for the efficacy of standard solutions.

While I don’t have too much in the way of specific recommendations, it might also be useful to think about how reform efforts interact with broader structures of patronage and power-sharing that define fragile states. Across-the-board system reforms are likely to mobilise broad resistance if they undermine the ingrained patronage and rent distribution systems that are vital to social order in fragile states. Reforms might be more successful if they are sufficiently targeted to avoid broad reaction from powerful interests, or if they can be combined with measures through which losers can be compensated and destabilisation of inter-elite power relations avoided.

Read the full paper here.

image_pdfDownload PDF

Tobias Haque

Tobias Haque has worked on economic development in fragile states for the past decade, and is currently a PhD student at the Department of Pacific Affairs at ANU and the World Bank's Senior Country Economist for Afghanistan. He holds Masters degrees in finance and sociology and undergraduate degrees in economics, political science, and development studies.

1 Comment

  • Thanks Tobias and congratulations for tackling a difficult political issue. Apart from the access to force or violence, the analysis of the lack of transmission of PFM reform efforts into service delivery outcome improvements is consistent with my most recent experience in Lao PDR. And the rent generating opportunities are multiplied in decentralised settings. The three guiding principles are useful, and although it may be difficult for donors and other stakeholders to explicitly accept the provision of compensation for the surrender of rent entitlements, there may be more nuanced means available to compensate the losers.

Leave a Comment