Surprise! Successful Community-Driven Development Requires Communities to Drive the Development!

By: Kara Weiss, Vice-Chair and Head of Programs at CRI Foundation & Stacey Faella, Executive Director at The Woodcock Foundation

It probably isn’t a coincidence that the generation responsible for igniting the civil rights movement in the 60s is the same generation that contributed to the then burgeoning community-driven development trend in international aid in the 90s. Ultimately, both have as their foundational presumption an extraordinarily basic idea: all people are inherently capable and should be treated as such.[1]

Community-driven development (CDD) is an approach to aid that puts the decision-making power directly in the hands of the community the grantor is trying to serve. As such, it’s up to the community to decide what they need — not the government, and not a non-profit. Nobel-Prize winning economist Angus Deaton, who is generally critical of foreign aid, suggests citizens are usually excluded from the decision-making process and therefore don’t amass any long-term benefits.[2]Anyone who has been to any corner (or frankly, right in the center) of any of the world’s poorest countries has seen this first-hand, even if they don’t know it. As an exercise of practical insanity (if insanity is defined as doing the same thing repeatedly and expecting a different result), the international community has spent trillions of dollars in aid to the developing world over the last sixty years or so and has achieved astoundingly little. More than a quarter of countries in sub-Saharan Africa are actually poorer than they were in 1960.

Writ large, we’ve failed. But that doesn’t mean every attempt has been a failure.

As Dr. Deaton points out, citizens are usually excluded from the decision-making process, and that exclusion causes the projects’ demise. But rather than drawing the conclusion that we shouldn’t attempt aid, why not address the disfunction in how aid is delivered: let’s include citizens. At its heart, this is what CDD does that is different from other types of aid. CDD is built on the philosophy that communities are themselves best suited to determine what they need, and, if provided sufficient resources, to create and sustain it. Real community inclusion leads to lasting results which leads to a return on investment.

No more crumbling wells. No more un-staffed hospitals. No more deserted schools.

When communities are armed with tools and resources to drive change for themselves community buy-in is established from the beginning. The conversation is never: Where do you want the school? Or even, what kind of school building do you need? But rather: What do you think your community needs most, and how do you want to get it? Through the process of CDD, communities learn how to self-organize, local governance improves, and as a result of community leadership, and ownership, the outcomes of CDD projects should be sustainable over time and drive long-term impact.

Lately, however, the narrative is being told with a different ending.

A new report, which attempts to synthesize the findings of twenty-three CDD programs, published by 3ie (International Initiative for Impact Evaluation) comes to the conclusion that “people may have participated in making bricks, not decisions.” And more specifically, “Many people may be aware of the programme and the community meeting, but few attend the meetings and fewer still speak or participate in decision-making.” The authors thus conclude that these CDD programs fail to lead to governmental and social impacts and that these objectives should be abandoned.

At the core of Community-driven Development is the requirement that the process be community-driven. If the evaluated programs were unsuccessful in sufficiently engaging the community such that most did not actively participate or feel a sense of ownership in the decision-making process and resulting project, then these programs may be local, but this is not the same as being community-driven. When community members don’t participate, we can’t expect a program to yield the benefits of participation. (Obviously.)

In one respect, we agree with the authors: Community-driven Development programs that don’t actually include the community should be eliminated. Because they aren’t community-driven. While we can’t pretend to guess whether the authors of this report intended to lead readers to conclude that CDD generally doesn’t work, that is exactly what they did. What we wish the authors had stated more directly is that they found a lack of rigorous attention to the key driver of success of CDD: community facilitation (which of course is not glamorous, not fast, and not easy to explain), thus resulting in a general failure of these programs. In fact, there were no criteria described for minimum acceptable levels of facilitation, no standards for participation and inclusion, and the reported lack of community participation signals to us that there really wasn’t a community-driven process taking place. Notably, the programs that did have increased facilitation and government engagement did indeed appear to have long-term impacts.

At this critical moment, we have to avoid the impulse to equate the reporting of failed programs with a failed idea. We are optimistic that funders and implementers will take stock and re-boot and dig deeper rather than walk away. It’s a test of nerves, but one we’ve seen win out in the health sector, with Community Health Worker (CHW) programs. Since the 1980s, implementers around the world have tried to scale successful place-based CHW programs and found that, time after time, they fail. And not only have they disappointed, they have failed fantastically. In response, six NGOs that have found success in their efforts to implement and scale their programs conducted a cross-organizational review of protocols. They found that, while there is much that distinguishes these programs from each other, there are eight discrete program design elements they all had in common that represent the minimum viable elements for CHWs to succeed. Moreover, programs that lack even one of these elements, are at high risk of failing almost entirely. As such, these design elements work as one cohesive requirement.[3]

The same can apparently be seen in CDD. If you aren’t able to galvanize and organize a community, you will not get the outcomes you want (i.e. community cohesion, civic action, sustainability, gender equality, increased hopefulness, etc.).

If the problem with CDD as it stands is that projects are not actually being driven by the community, what can we do? We can resolve to only support projects that are. What you’ll find is that not only does the ethos behind these projects feel better, but the outcome data actually is better.

If you take as an example the CDD program, Spark MicroGrants, you’ll find that Spark invests somewhere in the neighborhood of two times the amount of time and money into the facilitation process as traditional programs. As a result, Spark is able to avoid the primary and repeated shortcoming of traditional CDD programs highlighted in the report: the (dreaded) ‘funnel of attrition’. The funnel of attrition refers to a declining number of people involved in each stage of decision-making; a) knowing about the program, b) showing up to meetings, c) participating in meetings and d) participating in decision making. Moreover, the report observed that the funnel of attrition affected women more than men, stating: “women are half as likely as men to be aware of CDD programmes, even less likely to attend the community meetings and even more less likely to speak at the meetings”.[4]While this is a determining factor of how ‘community-driven’ a program is, shockingly only nine out of the twenty-three programs were able to produce data on community awareness and participation. This bears repeating: only 39% of the programs had data on community awareness and participation. (Perhaps this is a qualm for another day, but doesn’t it seem grossly negligent to fail to collect the key data that would indicate whether the required program outcome was being achieved?)

A community in Mbugayera, Rwanda deliberating and voting on the type of project they want to implement.

If we believe that participation is a key component of community-driven development, then we ought to know what it means to have participation. We would suggest that weekly to monthly community meetings are necessary, that a significant number of households should be present, that women should make up at least half of the attendees and that citizens should not only get to show up and voice their opinions, but have the decision making power over their village’s future.

Notably, there is no clear data on frequency of meetings, which could mean the difference of one to forty total engagements. This seems to indicate that programs are not meaningfully engaging citizens, or there was no interest in tracking whether citizens actually held decision making power vs. just voice or knowledge about the programs.

We must have dual outcome expectations, both that of projects that are relevant and meaningful to community members, such as schools, farms, and electricity lines, as well as outcomes in participation, community decision-making, and village savings, otherwise citizens continue to be used as tools in a distribution chain resulting in a product no one wants.

Furthermore, the data on gender inclusion is limited in the report. A summary note pointed out: “where female participation is a target, not a requirement… women are only half as likely as men to be aware of CDD programmes, even less likely to attend the community meetings and even more less likely to speak at the meetings.”[5]In contrast, Spark sees female participation consistently higher than male participation: 58% of attendees are women and 53% of participants are women.This is particularly significant given that traditionally male-dominated cultures in which Spark works.

Women in Bumasikye, Uganda taking a leadership role in the financial planning of their community project.

It is unclear from the summary if programs that do mandate participation lead to meaningful female participation. It was also noted that gender inclusion was not always listed as an outcome. We believe that first being clear on expectations of programs is important, and if something is to be community driven, there must be broad-based participation, which would include gender diversity and income diversity of participants. Once there is clarity on the expectations, there can be greater investment and deliberation on realizing the best approach to achieving it, such as using a quota system or a targeted facilitation mechanism.

If you parse the data rather than aggregating it, you see that CDD is a model that works when done right. When we make an effort to learn from both failures and successes, we can define the conditions and characteristics necessary for a CDD program to succeed:

  • High investment in facilitation and exacting attention on clarifying desired social outcomes
  • Provision of open village funding as opposed to a menu of project options
  • Commit funds upfront through a village grant as opposed to a competitive process for discrete projects
  • Weekly to monthly village meetings (structured and required)
  • Provision of decision-making rights to all (adult?) citizens rather than only to committee members or selected government officials
  • Have citizens elect mobilizers whose duty it is to ensure broad-based attendance
  • Gender mainstream the facilitation process
  • Have facilitation guidelines and tracking mechanisms

A number of these practices are tied to strong facilitation and an emphasis on diverse community participation. As the paper states, the first set of CDD programs focused on poverty reduction & infrastructure, then it transitioned to focus on decentralization and improving local governance and social cohesion. At the same time, programs with facilitation increased from half of programs to 100% of programs after 2000.[6]We believe the next wave will be about community capacity and civic action, and with facilitation being a key driver, funding for increased facilitation must follow.

If we learn from the story of Community Health Workers, we can come to understand the importance of getting CDD right in order to ensure communities finally have their right fulfilled to be in control of their own future. You don’t get 80% returns for putting in 80% of work — you get 10%. If you underfund the facilitation piece — and we know that facilitating community participation is a key driver of success — then we won’t realize the social impacts that matter.

Although dead privileged white men are not usually a particularly good north star for guiding thinking about international development, we will borrow from Abraham Lincoln this once: “government of the people, by the people, for the people.” If we uphold this ideal — if we believe that all people deserve equal opportunity and control in their own lives — we cannot write off an approach to programming that moves power closer to citizens. We cannot decide to pull funds out of a program objective when the most critical program design element was underinvested in, in the first place. Rather than dropping our ideals, we must invest in them through clear program objectives, funding, and evaluation efforts. We must stay the course, replicate what we know works, change what doesn’t, and continue to pursue a world in which all people are empowered to determine their own futures.

[1]Note that while we are rooting our ‘knowledge’ on CDD programs and literature since the 1990s, CDD draws from a long history of organizing and leadership in indigenous communities, emerging democracies and progressive policies in pockets around the world from Pakistan to the Amazon.

[2]Deaton, Angus. The Great Escape: Health, Wealth, and the Origins of Inequality. 2013

[3]Practitioner Expertise to Optimize Community Health Systems

[4]Community-driven development: does it build social cohesion or infrastructure? Page iv.

[5]Community-driven development: does it build social cohesion or infrastructure? Page iv.

[6]Community-driven development: does it build social cohesion or infrastructure? Page 4.