Solving the Last Mile Problem for Global Development

Why understanding the costs of reaching the most vulnerable people can lead to innovative solutions to serve them.

Solving the Last Mile Problem for Global Development

September 29, 2019

Deep in the Amazon, a boat chugs upriver with an ATM onboard to bring cash to the region’s most remote villages. In isolated settlements of Africa, drones deliver needed medical supplies. In the mazes that are the favelas of South America, satellites create geocodes in lieu of addresses so children’s births can be registered and they can enroll in school. These are some of the innovative ways that we are reaching the people who are invisible to the rest of the world.

But it’s a constant struggle. We know what we need to do. The question is not what to deliver but how to ensure that interventions reach and impact the hardest to reach, especially when the costs exceed political will.

The Reach Project, a research initiative at the University of Toronto made possible by a unique partnership between the Munk School of Global Affairs and the Mastercard Center for Inclusive Growth, is dedicated to understanding how to get the “important  stuff” to everyone who needs it. Now in its third year, the Reach Project helps scientists, policymakers and practitioners unpack the success stories for lessons to pass along. 

The work is anchored in the simple but often neglected premise that there are seldom any economies of scale in reaching the last mile in development. Indeed, the marginal cost, the additional resources required to reach the last 20, 10 or five percent of beneficiaries can be multiple times the average cost. In other words, ensuring that “no one is left behind” —the ultimate aspiration of the Sustainable Development Goals— can get increasingly and exponentially expensive.

Costs of Delivering the “Important Stuff”

Learning from Success Stories

It would be nice if we could ignore economics altogether, but the world is one of limited resources and shifting politics. When the costs get too high, political will tends to disappear, which means the poorest of the poor suffer the most. We must therefore find a way to triage the limited resources and for those that carry the highest marginal costs to find new ways to deliver services. We’ve been working on that at the Reach Project.

Most recently, at its second annual Symposium in Toronto, Canada, Reach Project researchers presented the results of 16 months of rigorous, on-the-ground analysis on various “last mile” projects:  how Sri Lanka eliminated malaria, how Ethiopia addressed food insecurity, how the Palestinian territories brought a cash transfer program to extremely vulnerable households and how Tunisia connected hard-to-reach populations to basic services via its neighborhood upgrading programs. 

The following four case studies distinguish between two types of interventions: those expanding basic human rights, such as access to food and education, and those delivering other public, economic or social goods, such as cash transfers and paved roads. We also distinguish interventions by their marginal costs to reach the last mile. 

Basics Human Rights with Declining Marginal Costs
Food Cash Transfers in Ethiopia
Food security is a basic human right.  Ethiopia’s Productive Safety Net Program maximized its reach of cash and food by allowing beneficiaries to choose their method of receipt (i.e., cash, direct food aid or a combination) based on efficiency. It also used pre-existing infrastructure and social services and relied on established third parties for distribution. 

Basics Human Rights with Increasing Marginal Costs
Indian Voters in Remote Mountain Regions
To reach voters in the Leh district in the northern mountains of India costs $1,600  per voter. One would need to take a plane, ride camels and hike overnight to reach these citizens. Yet, there is a moral obligation to provide access to political participation for the functioning of any liberal and representative democracy.    

Other Development Goods with Declining Marginal Costs
Social Service Delivery to Remote Areas in South Africa
The Integrated Community Registration Outreach Program (ICROP) provides mobile social services to South Africans living in remote areas. Mobile units (trucks) travel to hard-to-reach areas, equipped with internet connectivity. They provide integrated services, including birth registration and applications for various social security grants. By creating a one-stop-shop for government services, the South African government was able to significantly reduce the cost of reaching remote populations.

Other Development Goals with Increasing Marginal Costs
Understanding Informal Neighborhood Settlements in Tunisia
In upgrading informal settlements in Tunisia, the Tunisian Urban Rehabilitation and Renewal Agency put in place a neighborhood upgrading program. Unfortunately, it created a perverse incentive that encouraged citizens to continue to build informally, as they knew they would still receive program benefits.    

There is a compelling case for any services falling under declining marginal costs: as we include more beneficiaries, the intervention becomes cheaper. It is also clear that for those interventions with increasing marginal cost, particularly the ones that fall outside of basic human rights, we must double our efforts to find innovative ways to reach them and reduce the marginal costs.

In Tanzania, for example, delivering medicine to the “last mile” communities was extremely costly, and even more so during the rainy season when roads became impassable. Using the GPS data the trucks collected, consultants in Tanzania, working with government planners, were able to map the secondary road network across the entire country and classify all roads in terms of travel speeds and time to health facilities.  To speed things up even more, they created a mobile warehouse in the shape of a five-ton truck where smaller trucks would load up at different points along the routes, further lowering costs.

Similarly, “Community Pass,” a digital platform Mastercard is developing, brings many different entities under one proverbial roof to serve individuals. By signing in with a universal ID, people can access multiple services through one platform. A smallholder farmer has at her fingertips the ability to check in to her health care clinic and the health providers can match her information to vaccine records. She can connect her produce to more buyers or pay her children’s school fees remotely. Bringing additional services in one system dramatically lowers the marginal costs of the individual services and interventions.

Delivering development interventions to the very poor is expensive, not because of the price of the intervention, but because of what it costs to identify, find and deliver services to the poor and marginalized. Therefore we need to ask, is shipping ballots to remote islands (instead of pushing to implement electronic voting mechanisms that are safe) really the best use of limited resources? If not, then what is.

This reality demands the concession that not all important stuff is made equal. But that only spurs us on to work together, across sectors, to find new solutions to overcome this high hurdle—it is not the what, but the how.

About the Authors:

Arturo Franco is a development economist and VP, Data and Insights at the Center.

Joseph Wong is Professor at the University of Toronto and leads The Reach Project. 

Ali Schmidt-Fellner is Manager, Knowledge and Insights at the Center

Aditya Rau is an Analyst at the Center and a Managing Fellow at The Reach Project.