Does localization actually work? We look for evidence

Localization has been part of the mainstream aid reform agenda ever since the 2016 Grand Bargain put the issue firmly on the map. Proponents of localization such as the U.S. Agency for International Development outline two main strands. First, local entities should have more influence over what aid is spent on and how it is delivered. Second, development programs should be delivered by local governments and local organizations, which should therefore receive a larger share of the funding. While some advocates believe these changes are simply the right thing to do, others argue that there are also practical benefits. Localization is often cited as making aid more efficient and cost-effective. It is said to lead to more cash remaining in local economies. Some even say it could help to address the sector’s safeguarding challenges. But what is the evidence behind these claims? How much do we know about the impact of localization? The answer, perhaps, is not that much. While anecdotal evidence is plentiful, the formal evidence base remains relatively slight. There have certainly been attempts to document the impacts and benefits of localization. A 2022 analysis from The Share Trust, which supports grassroots initiatives, suggests that aid delivered by local groups is 32% more cost-efficient, by stripping out inflated international salary and overhead costs. There is also some formal evidence that recipients prefer to receive aid from local groups. A survey following the 2021 earthquake in Haiti found that local aid groups were perceived as more “honest” and “respectful,” while a Start Network study in Pakistan and Guatemala found that communities perceived local and national groups to have a better understanding of local contexts, to be more accessible, and to be more trusted, reliable and accountable. But experts say that, so far, relatively little attempt has been made to gather comprehensive evidence about the impact of localization — and how to implement it most effectively. As the Humanitarian Advisory Group, or HAG, an Australia-based partnership, observed last year in its “A Pathway To Localisation Impact: Laying the Foundations" report: “No systematic process has been undertaken to define the intended outcomes of localisation for crisis-affected populations, or to measure whether change is in fact happening.” It continued: “This doesn’t mean that localisation isn’t having an impact. … However, In the absence of consistent design, data collection and analysis ... we cannot be definitive about if, when and how localisation delivers better humanitarian outcomes for communities.” Speaking to Devex, John Bryant, a research fellow at the ODI think tank, agreed. “There is the assumption that localization will improve the quality and impact of humanitarian responses,” he said. But there is “very little … generalizable evidence to test those assumptions.” So why is this — and what does it mean for the localization agenda? One problem is definition. If you want to test a hypothesis, you need to know exactly what you’re testing. In the absence of an agreed definition of what localization is, or what the potential benefits might be, that becomes difficult. "Since you are not starting from a universally accepted definition of what localization is, there's straight away an issue in terms of measuring it,” Bryant said. “There is lots of anecdotal evidence around country level initiatives and country based pooled funds as equitable partnership practices, but whether any two organisations can look at those examples and say, ‘that's exactly localization as we understand it is,’ that really is a major problem." Another major problem is that there is little incentive to do this work, at least for the organizations implementing projects. “The most widely cited barrier to effective localisation impact measurement is that donors do not require partners to demonstrate change,” the HAG report said. The report quoted one interviewee who put it more succinctly. “If it’s not a donor requirement, of course organisations aren’t going to measure localisation,” they said. Bryant also noted that although many international agencies will record things such as the number of local NGOs that participated in a project, these are “proxy indicators” that don’t tell us anything about the degree of influence those organizations had or the impact of their inclusion. Beyond this, the HAG report identified that localization faced many of the same challenges that programs generally encounter when trying to assess impact, such as the need to measure outcomes over a long-time scale; the difficulty attributing a particular impact to localization; or the fact that the organizations gathering data are being paid to deliver a service and are therefore incentivized to collect evidence showing they were successful. The report outlines a detailed framework of “causal pathways,” including possible benefits of localization, and several testable hypotheses to assess whether localization is indeed delivering those benefits. Why do we need to know this? Some of the experts who spoke to Devex questioned whether measuring the impact of localization misses the point of the movement. As one U.N. consultant, who spoke on condition of anonymity due to professional sensitivities, said: “It is a core human right, the right to have a voice. You can't take rights and turn it into a measurable project.” But others disagreed. HAG, for example, gave examples that could cast doubt on whether local service delivery was always better. International organizations are sometimes perceived by communities as providing higher quality services, the report said. And in areas with strong sectarian divisions, international groups are more likely to be perceived as neutral. HAG said that evidence plays a vital role in convincing policymakers to adopt the approach, and in helping practitioners to understand when and how these approaches work best. Their report suggested that evidence is important not just to show that localization has a positive impact, but also to understand how localization should take place. “Organisations need to understand their impact (positive or negative) in order to inform decision-making about adopting localised approaches,” the report said. “This understanding will support more evidence-based localisation practice, ensuring it is relevant to specific contexts and enabling identification of where and how it can be strengthened.” In other words, even if there is a widespread consensus that localization is morally the right thing to do, evidence is still needed to help make it happen in practice.