Should the World Bank become more adaptive by weakening its safeguards?

September 27, 2017

Are grassroots faith organizations better at advocacy/making change happen?

September 27, 2017

Legitimacy: the dark matter of international development

September 27, 2017
empty image
empty image

Guest Post by Aoife McCullough, Research Fellow, ODIaoife-mccullough_jpg

Many donors work on the premise that a state can move from fragile to ‘stable’ if its legitimacy is strengthened. Accordingly, there’s a broad donor consensus that interventions in fragile states should include a mix of activities likely to contribute to increased state legitimacy – what the World Development Report 2011 calls ‘restoring confidence’.

In practice, the emphasis on ‘restoring confidence’ often translates into programmes that aim to increase access to services. The suggestion is that improved access to services will strengthen the state/society contract and, state legitimacy will follow. That is why millions of aid dollars have been spent on improving access to services in fragile states: In 2012, 45.4% of total ODA to 50 fragile states was spent on ‘economic foundations and services.’

As with most solutions in international development, it’s just not that straightforward. Here are three problems with the way international development actors currently seek to improve state legitimacy.

  1. Trying to ascertain how a state gains legitimacy is complicated

State legitimacy is based on what people consider as the right way to exercise authority – if the state broadly conforms to those expectations or ‘rightness’, then people will consider the state legitimate. Those expectations are not formed in a one-way process. People’s expectations are, in part, influenced by the history of the state and by the laws and principles that have been negotiated between the state and its citizens over time. That’s not an easy process to influence through an aid programme.

  1. There’s very little evidence for services improving state legitimacy

Recent large-scale cross-country quantitative and qualitative research carried out by the Secure Livelihoods Research Consortium (SLRC) over a period of six years shows that even influencing perceptions of government through service delivery is not straightforward.

legitimacy-is-the-elixir-of-political-power-quote-1SLRC asked almost 9,000 respondents in five countries (Pakistan, Nepal, Sri Lanka, Uganda and DRC) about access to and satisfaction with services, and about their perceptions of their government.  Following the first round of the survey that was carried out in 2012/2013, the SLRC went back to the same people in 2015/2016 and asked them the same questions, enabling a comparison between locations and two points in time. The surveys didn’t measure legitimacy per se, but perceptions of government – an important aspect of the construction of legitimacy – and therefore provide an indication of just how tenuous the link between access to services and state legitimacy is.

The research found that increased access to public services such as health care or education did not significantly change people’s perception of government, in either a positive or negative way. In the aid industry, there is often a worry that delivering services directly through NGOs will result in delegitimizing the national government. The findings from the surveys show that service delivery by non-government agents such as NGOs did not necessarily detract from people’s perceptions of government. What appears to be more important in influencing people’s perceptions of government is the way that services are delivered and experienced.

If people have access to grievance mechanisms or are aware of opportunities to influence decision-making about a service – through community meetings for example – their perceptions of government are more likely to be positive. However, the effect of having access to a grievance mechanism was not strong – in fact, the type of experience that had the most influence on perceptions of government was a negative experience.

This finding raises the uncomfortable possibility that delivering services that create problems for people may be more damaging to perceptions of government than not delivering any services. In fragile states, it is probably impossible to deliver a service that does not create a problem for service users in some form.

The survey also found that the wider political context influenced perceptions of government. In three of the countries, significant political transitions took place that had a clear impact on how people perceived their government. For example, in Sri Lanka, a northern-supported government was elected in 2015 and the percentage of respondents who thought that central government cared about their opinions grew from 44% to 65%. But as most aid programmes are not designed to facilitate significant political transitions, this finding highlights the extent to which an aid programme is likely to have a limited impact on state legitimacy.

3. Legitimacy is hard to pin down
The only time that state legitimacy becomes a tangible, identifiable phenomenon is when a state loses its legitimacy in a particularly visible way. It’s not until people begin to act on their disbelief in the state that its lack of legitimacy becomes clear.

Yet it would be wrong to assume that the absence of revolt indicates a legitimate state. The cost of revolting against a state is extremely high and many people living in states that no longer hold legitimacy simply choose the less costly option of cooperating.

If legitimacy is so hard to pin down, can donors meaningfully support its development?  Donors can use proxies – legitimacy 1like perceptions of government – to measure the impact of their activities on state legitimacy. But citizens’ perceptions of government are only part of the mosaic that makes state legitimacy. Claire McLoughlin’s suggestion of six ways in which services influence state legitimacy is a great overview of the wide range of processes that contribute to legitimacy. But it’s not clear how an aid programme would track impact on some of these processes. For example, one of the factors that influences how a service might influence state legitimacy is though what McLoughlin identifies as the state’s ‘legitimacy reservoir’, i.e. the foundations of legitimacy making in a state. Any effects that service delivery might have on the state’s legitimacy reservoir will be difficult to identify within the timeframe of most aid programmes. In any case, most aid programmes don’t have the capacity or budget to do the kind of monitoring and evaluation necessary to measure impact across such a diverse set of processes.

Perhaps it’s time to be less ambitious

Robert Blair first used the term ‘dark matter’ to describe legitimacy when attempting to measure it in a field experiment in Liberia. If defining and measuring legitimacy is proving too tricky for social scientists, maybe it’s time to admit that trying to increase state legitimacy with aid programmes is too ambitious. State legitimacy is an important concept and decision makers for development programmes need to be cognizant of the extent to which a state may be legitimate or not. But by treating legitimacy as an outcome that aid programmes can influence and measure impact on, the development industry is fooling itself.

There is a way that the development industry can engage more effectively on the issues of legitimacy. State legitimacy is a constantly evolving process that both state actors and non-state actors engage in to produce narratives about the state. An important function of these narratives is to justify who benefits from the state and who is excluded. Service delivery is one of these functions about which legitimating narratives are produced. For example, a legitimating narrative is currently being produced in the UK justifying why certain people need to pay for the NHS. Instead of aiming to increase state legitimacy through service delivery, the development industry would be better off focusing on increasing the legitimacy of inclusive service delivery in countries where they are working to improve service delivery. This is a sensitive process that needs to take into account overall narratives of state legitimacy. In fragile states, where a political settlement is contested with violence, it will be especially tricky. However, by focusing on legitimating narratives for inclusive service delivery, aid agencies can at least aim for something that is tangible and measurable.

And ODI asked me to plug two new papers on ‘fragile and conflict-affected situations': one on tracking change via panel surveys; one on ‘How to support statebuilding, service delivery and recovery‘.


  1. Thanks for these thoughts, Aoife. It is very relevant to me right now as I tie up the last bits of my thesis on health system development in the ‘fragile state’ of Myanmar [by the way, I ditch the term ‘fragile state’]. I conclude that, in some instances, INGOs are better to concern themselves with the effectiveness of their programmes to do what it says on the box i.e. “improving health outcomes for communities” rather than entangle themselves in how the programme might also engender state- and peace-building. I’ve observed the international aid community ditch some important cross-border programmes with ethnic civil society organisations in favour of working on state-building initiatives inside Myanmar, which seems counter-intuitive.

  2. Perhaps it’s time to move past intermediate concepts such as legitimacy to focus on the frequently forgotten bottom line of development: to improve the quality of people’s lives. This is not an argument for abandoning a focus on politics, power, political economy and related priorities and modes of analysis, for they are indeed crucial to improving that bottom line. Rather, it is an assertion that lots of wasted analysis, effort and funding goes into trying to aim for ephemeral, elusive intermediate goals when phenomena such as improved service delivery (if it improves people’s lives) are worthwhile ends in themselves. If strengthened legitimacy and the most elusive goal of all, nation-building, in fact are realized by advancing the bottom line, great. But respecting and prioritizing what people most immediately want and need in their lives is more realistic and effective – certainly more so than development professionals’ and perhaps political scientists’ fixations on their own favored theories and concepts.

    1. I agree that improved service delivery is a worthwhile goal in itself but a focus on this can also lead to problems. If our goal is improved service delivery, then donors would be better off directing funding to stable LICs where the possibility of achieving improvements is more likely than in fragile states. The idea that service delivery increases state legitimacy helped justify funding services in fragile state despite the increased costs and risks compared with stable LICs. This thinking also allowed service delivery programmes to be integrated into statebuilding and peacebuilding programmes enabling donors to spend on tangible activities, while remaining assured that they were contributing to the more abstract aims of statebuilding/peacebuilding. The point is that aid agencies always face difficult decisions about where and how to invest in improved service delivery so there is a need for overarching goals guiding decisions. Building state legitimacy is the wrong overarching goal guiding decisions about service delivery. Inclusive service delivery that wrestles with….in some places, very sensitively,….evolving political settlements, seems to me to be a good overarching goal. But I’m very open to other ideas!

  3. I totally agree with the comments above. Aid should reach the peiople its meant for and mske a difference to their life in frsgile environments. It gives the space to build its legitimacy which comes from many other factors varying in esch contexts

  4. Really interesting piece that brings two issues to my mind. 1) as others have said, sometimes a health project is…well just a health project. It speaks to the trend with some donors for somewhat ridiculous scope creep based on their current fads. It can be hard to placate a donor who wants you to include in your little health project pieces that can help boost stabilization, increase environmental resilience, increase awareness of disabilities, address GBV, etc. That scope creep can actually lead to a dilution of the quality of your core objectives which feeds into; 2) the article briefly mentions potential harm “This finding raises the uncomfortable possibility that delivering services that create problems for people may be more damaging to perceptions of government than not delivering any services.” So we’re saying that it seems aid programs don’t do much for legitimacy but are finding it can have the opposite effect when the services aren’t great…lots of food for thought here…

Leave a comment

Translate »