The Challenge of Localization

One of the most consistent themes of Samantha Power’s leadership at USAID has been her desire to shift decision-making power to local actors. In August 2022, she wrote that “[USAID’s] role is to support and catalyze local change efforts and processes.” In June 2024, she wrote that “local leadership is critical for transformative and sustained impact.” And she is not alone in her enthusiasm. In the UK, the Labour Party has repeatedly stressed the importance of “genuine partnerships” with developing countries, and the international development White Paper produced by the Conservative government Labour replaced committed to publishing a new localization strategy. Denmark, Sweden, and the Netherlands all (together with the UK and US) adopted the “locally led adaptation” principles at COP26. Donors have embraced the idea of localization. But the reality has proven trickier. In a new paper published today, I set out why, and what to do about it.

Unpacking “localization”

I identify four problems in turning talk into action:

First, we have yet to agree on what localization actually is. What do we localize? Who is local? There are many possible answers, which makes implementing something meaningful called “localization” difficult.

Second, we aren’t wholly clear why we’re localizing; and once we begin to think more clearly about it, it becomes apparent that localization has limits, and must change in character in different places.

Third, the desire to do more localization has not been matched by a change in the structures and processes adopted by donor agencies that have made the status quo not-localized (or at least, not localized enough, as evidenced by their stated objective of change).

Fourth, in addition to the practical barriers this mismatch creates, donors also face principal-agent problems internally and with their contractors to implement localization effectively. Understanding these problems better suggests ways of overcoming them.

The difficulty of pinning down exactly what localization means is already well-understood. While it may seem straightforward that localization of resources implies that more money is handled by local actors, and localization of agency means that more decisions are informed by or made by local actors, closer scrutiny shows that things are not so clear. What happens when the resource is knowledge, or technical assistance? Does localization imply that foreign experts or knowledge brokers answer to local agents, or that locally generated knowledge and expertise should be prioritized? It’s not obvious that one answer will always be better than the other. And when we localize agency, who counts as local? Is it just input from national bodies? Does input from village- or town-level bodies make decision-making “more” local? In some cases, it may make sense for local input to occur at a higher level, in part because the trade-offs made in determining the shape of development cooperation may mean making choices between very different kinds of support in very different places. USAID have responded to this difficulty by defining localization as the journey towards more locally led development processes; this eases, but does not eliminate, these issues.

But even once we’ve decided what localization looks like, we need to grapple with our ultimate motivation for doing more of it. I suggest there are three: because it is the right thing to do (i.e., for the intrinsic value of giving locals control of development processes); because it makes development cooperation more effective; and because it is politically important. These motivations do not march in lockstep. In particular, the first motivation includes a presumption that more localization is always better; and the last is opportunistic, to be pursued when there are political returns. But to maximise impact, localization will only sometimes be appropriate. There are times when the cost effectiveness or impact of an intervention is maximized when non-local actors are engaged (though it is hard to think of situations where some local engagement in decision-making is not useful). It will not always be maximized, and it should not always be pursued when politically expedient. Yet few donors appear to have seriously grappled with how best to evaluate when and how to localize, beyond—correctly—recognising that optimal localization is almost certainly greater than the very poor record of most donors today.

And once such work has been done, two more challenges await. The first is that donor procurement, project management, accounting, and reporting systems make it much harder for local organizations to meet donor requirements (and more so when each donor has different systems and standards), and indeed even discourage local organizations from bidding for contracts in the first place. And the second is that localizing effectively requires solving two separate principal-agent problems. One is within donor organizations, cascading the objective to change long-standing habits and familiar ways of doing things. One way around this is to set targets, as USAID has; another is to try and communicate a sense of mission around localizing—which in turn requires a sense of mission around its ultimate objective, most compellingly, impact.

The second arises when engaging with contractors and development actors. Contractors want to maximise the value of their contracts, while minimizing the costs they themselves incur. In implementing localization, donors will need to be wary both of international contractors trying to “look local” to win contracts, and indeed, if the pace of localization is too fast for effective local providers to keep up with, of truly local contractors without the capabilities required for effective delivery at the scale required.

Recommendations for better localization

Localization, then, will not be straightforward. I suggest five ways donors localize more effectively, and to better effect:

1. Define localization broadly for the purposes of strategy, but precisely in contracting arrangements. Doing so will allow flexibility in pursuing the right kind and extent of localization in different places, but—through precise definition in contracts—manage the risk of gaming by contractors.

2. Create a sense of mission around localization. Localizing effectively will entail substantial changes to existing donor practices. Achieving this requires staff at all levels (as well as existing contractors and parts of the delivery ecosystem) to change how they work. Targets can backfire; creating a sense of mission—around the impact that localization can support, for example—may be more difficult, but more rewarding.

3. Streamline and invest in project management and reporting. Though completely shifting aid modalities may well be more effective, in the medium term at least, it seems likely that most aid will continue to be project- or activity-based. At present, donor systems disadvantage local organizations relative to those well-versed in donor-country systems. Investing in reducing this disadvantage, through dedicated (perhaps local) intermediation platforms, for example, can reduce some of the practical difficulties of localization.

4. Re-engage local decision-makers. Most donors have allowed relationships with local decision-makers to weaken since the end of the general budget support era. While localization does not necessarily imply that donors need more local knowledge (with sufficient risk appetite they can also localize that by using local intermediaries), a better understanding of who makes decisions, who they represent and how different communities and constituencies in partner countries exercise voice is helpful.

5. Monitor and evaluate localization efforts. The evidence base on localization in practice is somewhat thin. Evaluation is needed to improve the process, to better understand its final impact and to learn more about how localization can proceed in different ways, and what the different costs, benefits and trade-offs different forms of localization have. Good localization strategies will include clear mechanisms for learning.

This final point is important. Localization is new territory for most donors, and certainly a departure from recent practice. There will almost certainly be unintended and unanticipated consequences—some good, and some bad. Learning about these consequences should be a priority.