The Aid Transparency Initiatives and the focus on the use of country systems, emphasized in the Paris Declaration, encourage donors to publish what they fund and to use existing country public financial management systems. However, this focus on the how of development assistance somewhat distracts from the what. The bigger question really is why donors and governments focus on those particular areas and why those donors are the right partners to begin with.
Much of the shape of development assistance stems from the interest that donor agencies have in offering projects designed around the skills and the experiences they already have “in stock”. The most striking example of this was in the dashed hopes for Poverty Reduction Strategy Papers (PRSPs) launched by the World Bank and the IMF in 1999 as part of the Highly Indebted Poor Countries (HIPC) initiative. Broadly, the idea was that through an extended participatory process, governments would set the agenda for development policy along more consensus-based lines. PRSPs were to address four main topics: (1) macro and structural policies for growth; (2) governance with emphasis on public sector financial management; (3) sectoral policies; and (4) costing of and funding for major programs. The technical expertise of the International Financial Institutions was to be balanced by the local knowledge of the key actors – donors, were to be "brokers of participation”, not overt dictators of policy options".
In reality, PRSPs looked strongly alike, regardless of the context. Suspicions of donor-funded consultants writing the papers for government abounded.
In sum, the picture of development assistance in general, and of assistance concerning Governance and Public Sector Management in particular, is that projects are proposed with limited opportunity for governments or others to raise questions about whether this is a priority area, whether the proposed type of intervention is likely to lead to improvement and whether the donor in question has the relevant track record.
Moving from ex ante claims to ex post accountability
Arguments have been made for some decades that the key problem with development assistance is weak governance of the aid system. The argument is mainly made in terms of the costs of the overlap between donors, but there are also concerns about wastefulness from multiple assessments and interventions concerning public financial management.
However, arguably, no governance arrangements for the aid system will be strong enough to counter the incentives of individual donors to “persuade” governments to take their assistance - through expertise (claimed or proven) or through the incentives which are built into aid projects (publicity, access to international career opportunities, local project resources). The urgency of that drive to “sell” projects likely arises in some degree from the entrenched interests within donor agencies – if they employ a large group of technical experts, how can they avoid developing recommendations in line with the professional views of that group – and is it likely that those professional views will be completely detached from the career interests of the staff involved?
The lack of challenge to donor proposals provides an environment in which donors learn little about governance and public sector management interventions because they are paid for ex ante opinions rather than ex post results. Limited public debate and contestation leaves donor organizations with a form of “monopoly expertise (which) tends to produce a poorer epistemic performance than competition” (Koppl, 2012, p.172).
This is not an argument against expert knowledge being developed within donor organizations – it is however, to draw a distinction between the way in which experts are used in the development community, primarily as a source of opinions on what will work, and how experts are used in other sectors where they are used as a source of testable solutions. A racecar driver or an entrepreneur has expertise, but their challenge is to apply it so that they “win”. A development expert also has expertise, but their challenge is the lesser one of showing that they have it through setting out theories and expert opinions.
This is not specific to the development community or to the governance and public sector management subset. “Expert failure” arising from lack of contestation of ideas and between experts has been consistently identified in the case of the criminal justice sector when the system of prosecutor, police and crime laboratory produce little challenge and result in “low error correcting power” and hence erroneous prosecutions. Much has been written about the imperviousness of expert judgments in the development field to hard evidence.
Help from the market?
Could a market-based solution be the reform to the aid governance system that makes the difference? Some elements of a more market-like environment are in place. As with other areas of development assistance, there are an increasing number of providers of advice on governance and public sector management. Just a decade or so ago, development assistance was overwhelmingly provided by traditional bilateral and multilateral donors, which provided aid in specific ways and according to a particular set of norms. Today, this is being complemented by the growth of other forms of development assistance, including from non-Development Assistance Committee (DAC) donors, climate finance funds, social impact investors, philanthropists and global funds, as well as less concessional flows. Overall, development assistance flows grew substantially over this period, with the bulk of the additional funds coming from donors outside of the OECD DAC, philanthropy, social impact investments, global funds, funds for climate finance, and, non-concessional financial flows from the development banks including the World Bank. In addition, some of the non-traditional donors, China in particular, have considerable technical assistance embedded within other infrastructure projects.
The consequence is that for support to governance and public sector management, along with most other areas of development assistance, governments have access to considerable, generally overlapping, advice. (A recent study has highlighted that in Vietnam has a ratio of 1 project per 9,000 people).
However, in order to allow governments and other commentators to participate in an active debate about which governance and public sector management aid product best fits country needs and which provider can best deliver it, and then to make a choice, then there is a significant information problem which must be solved. The effectiveness of assistance in this field ultimately can only be judged by the results that it brings. The track record in the field suggests that advance assurances that Governance and Public Sector Management projects will make a large and lasting difference are not particularly credible.
Thus donor aid product proposals must be accompanied by information about the evidence base on which the proposal is made. This would be in addition to information about costs (financial and transactional) and likely returns. This information about the evidence base would require, in essence, that project proposals are accompanied both by evidence from cross-country and multi-sector studies about how this type of reform proposal has worked out in general and by evidence from Impact Evaluation (IE) or other more finely grained research which can be adjusted to test for the salience of distinctive environmental conditions and thus able to answer the parallel question: “why is this likely to work here?”.
Where is the independent advice for governments?
Any steps forward require that donors accompany their proposals for assistance with sufficient information for governments to ask critical questions concerning the costs to them, the results that they are likely to see, and the evidence base on which the proposal has been developed.
This more market-based approach must start from setting stretching but feasible targets for development outcomes. Ideally this would draw on comparative empirics and analyses concerning improvements in similar contexts – but the essential question would be “we want to achieve this result within this time period, do you have a proposal which can help us do it?” However, emphasizing contestability and evidence in reform proposals raises the question about whether there is a separate line of assistance that governments need in order to become discriminating consumers of development advice. Given what seems to be donor ingenuity as packaging their offerings as the exact solution that the country needs at this particular juncture, how is such diagnostic guidance to governments to be provided, helping them identify the binding constraints to development without undue contamination from the supply of technical assistance from the donors.
One consequence of more contestability in reform proposals is that donor agencies would feel the pressure to develop cost-effective diagnostic approaches for understanding the challenges involved in meeting those targets – they would need some way to work out what they needed to deliver and whether they had the capacity – avoiding committing to help meet targets for results where either they do not have a comparative advantage or where they doubt that the results are attainable. One might hope that these forces might assist in moving the focus of donor management attention from projects to problems. Current management systems encourage donors to look through the telescope from project design through to implementation and attempt to find the most significant development objective that they can achieve with that project. Looking through the telescope from the other development objective end would focus managerial attention on the problem and allow more flexibility in the projects and other solutions that they bring to bear to solve it.