Lauren Abel and Andrew Blackman, recent graduates of the Harvard Kennedy School’s Masters Program in International Development, contribute the following guest post:
The Ebola virus spreading through West Africa has reached Nigeria. While the number of Ebola cases in Nigeria is small, the highly contagious disease can spread quickly, especially in a country with chronically low-quality health services. One might think that the biggest economy in Africa should have the resources and infrastructure to battle the health threat, but despite billions of dollars in annual oil revenues, Nigeria’s poor health services are putting the country at risk of spreading the epidemic. Unfortunately, the problems with public service delivery in Nigeria do not stop with health. Authorities have recognized that Nigeria is unlikely to achieve any of the Millennium Development Goals (MDGs) by 2015 (see the Government’s full review here).
Why the dearth of public services in a booming economy? Corruption. Above all else, pervasive and endemic corruption remains the Achilles Heel of the Nigerian economy. As the Finance Minister recently noted, “Corruption is a serious issue for us because it is destroying our country, eating deep into the fabrics of the economy, we can’t have infrastructure and development with these level of corruption.” In an environment of such endemic corruption, what can be done to help translate huge natural resource revenues into tangible improvements in the lives of the 110 million Nigerians living on less than US$1.25 a day?
In a recent study conducted with the Centre for the Studies of the Economies of Africa – an Abuja-based economic think-tank – we put forward an approach designed to improve the provision of public goods and services that we believe could work within the current Nigerian system. That approach is known as Results-Based Financing (RBF). (For an example of the RBF approach, see this World Bank initiative; the approach is conceptually similar to other ideas such as Pay for Performance, Cash on Delivery Aid and Output-Based Aid.)
To understand how RBF would work in Nigeria, it is important first to understand that Nigeria, like many large countries, practices a form of “fiscal federalism,” in which the national government and the subnational governments (states and localities) share responsibility for public service provision. In the Nigerian system, the central government is responsible for collecting most revenues and for distributing these revenues to state and local governments. These subnational authorities depend on the center for revenues, but remain highly autonomous in their spending decisions. Weak political and administrative structures at these levels mean state governors are all-powerful and direct subnational funds as they see fit.
In the traditional approach to public service financing, revenue allocations from the center to subnational governments are based on the “inputs” that the subnational governments plan to procure – things like building hospitals or hiring health workers. The hope and expectation is that supplying these inputs will lead to improved outcomes. But input-based financing schemes are ripe for graft. Politicians can easily direct funds to preferred contractors who can use a fraction of the funds to meet minimum requirements, while pocketing the rest. Even when funds are used properly, international experience has shown that interventions focused on inputs do not necessarily led to better outcomes.
RBF, in contrast, directly links central government financing with the monitoring and verification of results, such as a reduction in maternal mortality or an increase in the rate of infant immunization. By linking payments with measureable outcomes, RBF makes it harder for corrupt politicians/officials/providers to cheat the system, since real results must be achieved and verified before funds are disbursed. Additionally, RBF is designed to work within existing government structures, and hence to strengthen transparency and local accountability by encouraging community participation in delivery and monitoring of service provision.
How would an RBF-style program work in Nigeria?
Our RBF proposal is designed to provide conditional federal financing to subnational governments to improve specific human development outcomes within their jurisdiction. By offering federal financial and technical support, the program would encourage subnational governments to tackle locally-nominated problems. Since the RBF scheme will be based on outcomes, the initiative would require a non-manipulable method for setting targets and for measuring outcomes; this method must be sufficiently ambitious but still feasible. To set these targets, we propose a technique that has been used to determine poverty reduction targets in Latin America and is discussed at length in our paper. (That technique has the advantage of being both transparent and easy to calculate – meaning it can be maintained by a bureaucracy with limited technical capacity.)
Furthermore, rather than offer a standard RBF package to all subnational governments, we instead propose that the central government offer the subnational governments a range of options that vary in terms of local autonomy, openness of the monitoring and evaluation (M&E) process, and financial risk. The subnational governments could then opt into whichever package is most appropriate. For example, better performing subnational governments that want more autonomy could opt for a design with fewer compliance requirements, but with a more open M&E process, and greater financial risk. In the paper, we provide an example involving three options, and show how the program can be designed to encourage subnational authorities to select the option that is most suited to their needs and capacities.
The RBF proposal will not solve all of Nigeria’s woes, nor does it directly attack the overall environment of corruption. It is, however, a pragmatic method to promote more efficient public service delivery, with less incentive or opportunity for diversion of funds. We believe the RBF approach provides a valuable alternative in the toolkit of policymakers in Nigeria, and in other countries where corruption undermines the capacity of the government to do what governments are supposed to do – create a better future for all citizens.
GAB | The Global Anticorruption Blog 
There are a number of intriguing insights in this study that merit further discussion but I’d like to share a quick reaction. Results-Based Financing depends, of course, on the reporting of accurate results. As the authors recognize, a robust system of monitoring and evaluation plays a critical role.
Having spent time observing large-scale M&E processes on the border between Borno state, Nigeria and the Far North region, Cameroon, I would caution against underestimating the risk and frequency of manipulated reporting of indicators. Even if data collectors and auditors are not paid off with outright bribes (hardly unlikely in an endemically corrupt environment), the target population will quickly learn that money is tied up in what they say. They may well adjust their reporting accordingly – an outcome that is particularly likely under a “high-profile federal program.”
Therefore, I think that community-based M&E is actually most important in marginalized “Class C” areas, where the sudden presence of external monitors would draw widespread attention and cause significant disruption. Given the language barriers and village idiosyncrasies that distinguish many of the neediest communities in northern Nigeria, I am unconvinced that outside monitors could effectively perform their jobs anyway. The strategy may seem to weaken oversight but I believe it would lead to more reliable results reporting.
As the study emphasizes, existing community resources ought to figure prominently in proposals for reform. Apropos to the post’s opening reference to Ebola – international, public health organizations operating in northern Cameroon and Nigeria rely very heavily on a network of “animateurs” residing in the villages they service. These community representatives would be good starting points for reformers seeking trustworthy, accurate documentation of the status of social services in underserved areas.
I agree with Elizabeth that monitoring the results accurately would prove extremely difficult (as well, sometimes, as monitoring the quality of the results; if the metric is number of schools built, for example, that would not take into account the quality of construction and likelihood of falling down).
I am also curious about the timing on the funding. In order to work, it seems like the local governments would have to pay the money up front to build the improvements and only be reimbursed from the federal government if they produce results. Do local governments have the cash flow to be able to take on a significant project without prior funding and to risk not getting paid at all if it somehow fails due to circumstances beyond their control?
Also, what happens when there are metrics we care about (say age of marriage for girls) that would be likely to change slowly over years as attitudes and opportunities change rather than something like deaths in childbirth, which would potentially change much more quickly with better healthcare access. Perhaps this system would only be useful for projects with the kinds of metrics that have easily measurable significant results within a year or so.
Overall I am very intrigued by the idea of Results Based Financing, but I think we need to figure out ways to make the results themselves much harder to tamper with.
Elizabeth and Sarah, thank you for your thoughtful comments. Yours are clearly important points. Let me have a go at responding.
First, I completely agree that involving communities in the M&E process is crucial (in fact, I would say that we WANT local people to know money is tied up in what they say, and thus they have a crucial role to play in ensuring that they get what they have been promised by their elected officials). Consequently, our proposal is to have a cadre of local M&E officers, who would work closely with the community, and local and state authorities, but who also have a direct link to provide monitoring and feedback to the state and federal RBF governance units. Under funding options that require stricter process-compliance (i.e. Option C in the paper), this regular, locally-led M&E process could be complemented by periodic federally-contracted independent verification. (Ondotimi Songi makes another valid point about M&E below).
Second, regarding the timing of payment. As we outline in the paper, the financial instrument would include 3 ‘prongs’: (i) a federal grant; (ii) a convertible federal loan; and (iii) local financing. All three would be available to subnational authorities immediately when the RBF project is approved. The ‘results-based’ conditional part only kicks in when the project is evaluated. If the project is evaluated to have achieved its objectives, then the federal loan is converted to a grant, with no liability on the subnational authority. However, if the project is not completed to the specifications outlined in the approved project proposal, then the Federal Government can request a refund of their share of project financing – which then does become a liability on the subnational authority. Clearly there is some degree of financial risk for subnational authorities here – which is why we propose having a ‘menu of options’ with varying degrees of conditional financing, which subnational authorities can select into. It also provides very strong incentives for subnational authorities to promise what they can deliver, and deliver what they promise.
Finally, I would not suggest that one program design would be suitable for all the objectives we care about. As you say, Sarah, the impact trajectories of some outcomes are intrinsically non-linear, and would require considerably longer than one fiscal year to see impact (i.e. women’s empowerment, FGM, AIDS awareness). This is not to say a RBF-style program could not be used for these types of objectives – only that the program design would need to be different from those objectives that we believe have a shorter impact trajectory.
I am wondering if RBF is compatible with Nigeria’s legal framework where federating states are autonomous politically and financially. The Federal government in Nigeria cannot dictate to the federating states how monthly allocated revenues should be utilised. Federal government in my understanding cannot seek to undertake a M&E exercise for projects undertaken by federating states as this still amounts to interference in the affairs of the federating states. It may be possible to propose the RBF framework at the state and local governments level where the latter are controlled by the former but as observed by Elizabeth, the RBF implementation in Nigeria still poses a lot of practical challenges in a country riddled with so much corruption.
Thank you for the comment, Ondotimi. I agree there are many practical challenges – that is where a process of iterative experimentation, learning and adaptation will be crucial to fit the program within the local context. Nigerian policymakers and implementers are clearly crucial to this iterative process.
I agree that the Federal Government cannot dictate to subnational authorities the allocation of spending. The need to work in partnership with subnational authorities, rather than dictate terms or independently administer programs from the center, are key lessons from past experience in Nigerian Federal Government programs (i.e. in education, health and water, which we discuss in the paper). Thus, the RBF program is designed to offer federal money as co-financing for local priorities – determined and designed by local communities and authorities. As we highlight in the paper, participation in the program by subnational governments is voluntary, and the decision of the funding option is autonomous. This provides state and local government with the flexibility to leverage Federal financing to achieve their own priority objectives (within a range of defined sectors, such as health, education, water & sanitation and local infrastructure projects).
As for the M&E process, this is also an important point. I agree the Federal Government cannot evaluate independent subnational projects. However, there is a precedent of joint Federal-subnational M&E of co-financed projects under the Federal Government’s Millennium Development Goals Conditional Grants Scheme (see chapter 6 of the CGS Implementation Manual, http://www.sparc-nigeria.com/RC/files/4.2.8_MDGs_Conditional_grant_scheme_implementation_manual.pdf). There is also precedent of peer-review by Federal and State counterparts under the State Peer Review Mechanism (SPRM), as part of the Nigeria Governors’ Forum (see http://www.nggovernorsforum.org/policy/state-peer-review-mechanism/). The peer-review element has been effective in raising the legitimacy of the reviews in the eyes of Governors, while also fostering competition between states to improve performance.