This year, we are running a small pilot in Uganda’s Acholi sub-region, on the border with South Sudan. Until just over ten years ago, this area was the heartland of Joseph Kony’s Lord’s Resistance Army, and the region was embroiled in a protracted civil war. Today, Acholi is at peace, but its people remain amongst the poorest in the country, living in some of its remotest areas.

Cash is now widely recognized as a uniquely efficient and effective means of alleviating chronic poverty. The evidence is compelling, from a randomized controlled trial (RCT) of our own program in 2013 to ODI’s comprehensive review of 165 different studies, published last year.

The development sector is taking notice. In Foreign Affairs, our founders, Michael Faye and Paul Niehaus, noted that cash is increasingly considered the benchmark for aid and development spending. The heads of the World Bank and the UN have both signaled their intent to increase cash programming, and the UK Prime Minister recently publicly recognized cash’s positive impact. However, cash remains a tiny proportion of overall development spending, at just 6% of all humanitarian aid.

Image not found
Recipients in Uganda gather for SIM card registration.

A significant barrier to the use of cash is the belief that it is not a viable intervention in remote locations, where disproportionate poverty is trumped by perceived operational complexity. Our pilot in Uganda is designed as a test of these operational challenges. If it is a success, our goal is to encourage other implementers to invest more in cash transfers in remote locations. We believe providing this case-study is an important first step in influencing other organizations on this subject. This project, and the learnings we draw from it, will also influence our own programs, as we continue to expand to new and increasingly remote locations around the world.

To provide this case study, we have chosen a location that fits multiple criteria of remoteness and resultant operational complexity. The location is Lamwo District, in the Acholi sub-region of Uganda, which:

  • Is physically remote – with 30% of our target households only accessible on foot, some 4 to 7 hours walk from the nearest passable road, and all households 61 to 85km from the nearest major town.
  • Has limited access to financial services – with a single mobile money agent covering an average of 254 km2, 39 times more than the average for mobile money markets as mapped by the Gates Foundation.
  • Has minimal cell coverage – ranging from intermittent and limited to nonexistent across our area of operation.
  • Has low population density – home to 24 people per km2, seven-times below the national average, in villages around a third as populous as our other areas of operation in Uganda.
  • Is extremely poor – the third poorest of Uganda’s 111 districts, according to government census data.
Image not found
Courtesy of, we are operating within the red circle in Northern Uganda.

Here, we intend to deliver around $500,000 to 500 households, over the course of 3 months, and measure our ability to do so. There are undoubtedly many questions to be explored with regards to cash in remote places – from a robust analysis of individual, market and community impact, to transfer sizing and timing. Remoteness is also sometimes paired with additional complexity, such as recent / ongoing conflict or food insecurity. We do not set out on this project purporting to provide robust answers to all of these questions. However, there are some simple but important questions we will be able to answer, including:

  • Can cash transfers be delivered and collected by recipients?
  • What is the operational cost to the implementing organization of doing so?
  • What quality of service can we provide to recipients in these locations?
  • How will individuals and communities respond to receiving cash, and what will they spend their transfers on?
  • How do nascent financial services react to an injection of cash?

It should be noted that we don’t take a positive outcome to any of the above as a given, even as fundamental a question as whether our recipients will be able to collect their cash. We expect that in Lamwo, as in our other areas of operation, the overwhelming majority of our recipients will seek to convert their digital cash transfer into hard currency the moment they receive it. To do so, they will need to ‘cash-out’ with a mobile money agent. While we are training recipients on the basics of how to use mobile money, we are taking a decidedly light-touch and organic approach to ensuring that they are able to ‘cash-out’ – similar to how we work in our other, less remote, areas of operation. Once we have sent the transfer to a recipient’s phone, it will be largely on them, and the responsiveness of the market, to convert it into cash. Should this not work here, and there is a real possibility that it may not, we will need to correct our approach, investing more heavily in the process of pairing our recipients with the mobile money agents and banks who can provide the required liquidity.

It is through experiment and iteration like this, that we will seek to answer the three simple questions that led us to this project: a) can cash be delivered to recipients in remote places?, b) how can it be achieved?, and c) at what cost to the organization who delivers it? Using these findings, we intend to build an evidence base that will influence how other organizations think and act in similar situations. In Sub-Saharan Africa, where 62% of the population live in rural areas (and a high proportion of those in inaccessible ones), we believe the potential impact of this small program could be considerable. In light of this, we will be sharing our findings with key implementing organizations in May/June 2017, and of course with you, right here.

If you have any questions, please do get in touch and continue the conversation. You can either email us directly at or tweet @GiveDirectly.


Josh Williams is a Field Director at GiveDirectly, based in Uganda. He is currently leading our remote payments project.

Back to List