Service delivery and state-building: the 46.7 billion dollar question?
Since 2007, OECD DAC members have been signed up to state-building as the ‘central objective’ of their engagement in fragile and conflict-affected situations. It is a compelling narrative: deliver basic services (health, education water and sanitation) and – voilà! – better prospects for long-term peace and stability. It is also one which has rapidly been translated into programming: in 2009, official development assistance (ODA) to countries described as ‘fragile states’ totalled $46.7 billion, with 43.7% of this going to social infrastructure and services.
But, while all this spending may help to achieve vital gains in health, nutrition, education and access to improved water sources, do we know if it is strengthening the capacity of conflict-affected states to perform core functions? What about the question of whether these aid flows are enhancing the perceived legitimacy of those states, or if they are generating positive changes in state-citizen relations?
In our review of the evidence on social protection and basic services in conflict-affected situations, carried out during the ‘inception’ year of the Secure Livelihoods Research Consortium (SLRC), we find that most claims made about contributions to state legitimacy and state-building are not backed up by empirical evidence (full review - Social protection and basic services in fragile and conflict-affected situations and Briefing Paper - Social protection and basic services in conflict-affected situations: what do we know?). In particular, we found only a very limited number of studies in which evidence, rather than assumption, is used to show how delivering basic services can contribute to state-building. And even fewer providing measurable robust indicators of state-building effects.
This lack of evidence is particularly problematic when (the received wisdom that service delivery contributes to) state-building becomes the central objective of programming. Development outcomes take a back seat to governance gains and, in some cases, this has implications for the geographical or social targeting of expenditures. In the Democratic Republic of the Congo (DRC), for example, USAID has shifted health programmes away from critically underserved areas in an effort to consolidate the fledgling peace process. Using health services simply as a means to an end (state-building), rather than an end in themselves (health outcomes that are good for poor people), poses challenging questions for the humanitarian principles espoused by the agencies implementing programmes on-the-ground for USAID and others.
At the same time, when service delivery programmes adopt state-building objectives without strong evidence that they will work, they become, to borrow from LSE’s Stuart Gordon, a 'house built on sand'. They risk not only failing to achieve their objectives, but also threaten to undermine support for aid in conflict-affected situations, support for investments in basic services in particular, and even support for aid in general.
What evidence we have found shows that the relationship between service delivery and state-building is complex, and critically in need of further and better research. Three issues in particular matter.
The first is national versus local. People experience the state at different levels and can have different views of each. Our paper reflects the importance of understanding that different or ‘hybrid’ systems of political authority exist (see Mick Moore’s recent post), and that we need a much more nuanced or context-based view of the ‘state’. The implications for programming are two-fold. First, policy-makers may need to rethink state-building objectives in places such as eastern DRC where the state is barely present and where the few basic services that exist are provided mostly by the church and NGOs. Second, a strong argument emerges that we must learn from the divergent experiences of Somaliland and Somalia, where governance systems (including those through which social protection and basic services are provided) have emerged at the local-level, from the bottom-up.
The second issue is supply versus demand side approaches to state-building. Emerging evidence from some conflict-affected situations suggests there is a case for an increased focus on voice, accountability and demand in basic service delivery – what could be termed a shift from state-building to ‘citizenship-building’. The main implication here is that state-building efforts should not simply support the capacity of weak governments, but also deploy participatory mechanisms to support democratic demand. This can, in turn, strengthen government capacity (albeit in full knowledge of the limitations often posed by exclusionary political settlements and neo-patrimonial regimes). Such a line of thinking resonates strongly with the findings of ten years of research at the Development Research Centre on Citizenship, Participation and Accountability.
The third issue is qualitative versus quantitativeevidence.So, we know we need more and better evidence. But what sort of evidence? Our analysis suggests that we need to move beyond the superficial binary of randomised control trials (which may fail to pick up other important determinants of legitimacy) and deep, qualitative analyses focused on context and texture (which may produce findings with limited transferability). As we have argued elsewhere, cutting out the ‘noise’ is not necessarily desirable in conflict-affected situations, where context and process have as much influence on programming outcomes as the choice of intervention itself. However, greater methodological diversity and rigour are clearly required in research on conflict-affected situations, and review-based approaches must not be a substitute for primary research, whether quantitative, qualitative or mixed methods.
These are serious challenges, but there’s enormous pressure to get this right. There are 46.7 billion dollars – and millions of vulnerable people – at stake.
This post features the author's personal view and does not represent the view of ODI.