Transparency is Insufficient: Lessons From Civic Technology for Anticorruption

Over a decade since the United States spearheaded the creation of the Open Government Partnership–a platform that includes over 74 countries, 150 local governments, and thousands of civil society organizations to promote transparent, participatory, inclusive, and accountable government–and the subsequent publication of terabytes of open government data by dozens of governments around the world, why do we feel like government corruption still thrives and gains ground even in developed countries? Why have increased transparency, access to public data, and even the spread of smartphones not led us to the world many envisioned technology would make possible early during the Arab Spring in 2011? Can civic technology really realign the relationship between citizens and governments in ways that promote integrity and trust?

To the extent that civic technology has failed, it has not been because of insufficient data, but because it often ignores power and collective action.

I spent the last 15 years working at the intersection of civic technology, democracy, and anti-corruption. In that time, I led the design and development of two software applications that tackled fundamental challenges for the education and healthcare sectors of my country, Paraguay, infamous for its high corruption levels. FOCO, an interactive map that makes around USD 360 million in annual funds for school infrastructure and meals transparent covering all public schools in the country, continues to drive local impact and receive international awards. Many years of trial and error, of idealism forced into pragmatism, and an intense focus on grassroots collective action have given me some insight into the questions above that could be useful for people envisioning and developing the tools that could update digital civic infrastructure for a renovated democracy while rebuilding trust in government and, as a byproduct, reducing corruption.

Civic technology starts, well, with the citizen bit before the technology bit. Too often many exceptionally sophisticated initiatives start the other way around, with the technology instead of the people. There is a large digital cemetery of fascinating applications—dashboards, automated red-flag systems, report apps, interactive maps, you name it—that most likely resulted from brilliant people who knew about the right group of datasets and tools that needed to be combined in order to shine light onto a problem, thereby allowing the potential for tackling it. However, as a result of prioritizing technical capabilities over user capacities, (i) the “build it and they shall come” narrative crumbles; and (ii) the level of process automation is such that citizens assume “someone” will do something, so they have little incentive to do something themselves.

I believe these problematic scenarios result from the lack of three things. First, there should be a problem definition centered around the people affected. If the problem is defined as “mismanagement of the public education funds,” the proposed solution will likely become an optimization task that is tech-centered; whereas defining the problem as “the education funds for infrastructure and school meals are not reaching the students in the neediest schools” would provide solution options that revolve around the students. Second, with these technologies there is seldom a process map of the problem that has the citizen at its center or end-point. The lack of said map seduces us to leapfrog the messy “human components and dynamics” to narrow down on the datasets that need to be published and connected. Third, there is seldom a theory of power regarding the way change happens. Depending on their design and use, technology can increase power asymmetries or balance them. Namely, when technology-enabled transparency increases along with impunity, citizens feel even more powerless. Using technology to show people the massive amounts of school funds that annually disappear due to corruption without facilitating actions to tackle this problem will certainly backfire. In cases such as these, technology is inadvertently increasing the power of the corrupt by making average citizens feel (more) helpless.

These three factors are often considered as second-tier items. In my experience, the lower importance assigned to them comes from avoidance of what focusing on them would expose: a level of social and political entanglements that would reduce the appeal of proposing technology as the go-to solution in the first place. Technology is necessary, but in my experience effective civic technology is usually quite simple, condensing complexity to a level that is actionable and intuitive for citizens. This is relevant both to actors in civil society and reform-minded public servants in government. Let me expand on why a citizen-centered process map and a theory of power are relevant. The former raises questions about intermediation and the latter about collective action.

Design from the citizen backwards

When the citizen is the end-point or center of a process, the layers of intermediation become evident. Sometimes these layers are parallel government agencies and their jurisdictions that intersect at random moments, siloes of datasets owned and collected by different entities, and the unclear dynamics that give life to the policy problem. What is more, the citizen might understand there is a problem (e.g. the money for new classrooms and school meals is mismanaged) but ignore how an interactive map of their schools’ resources could help solve it. Indeed, the citizen might blame the wrong agency or level of government. Ten years ago, when we first developed FOCO, we learned that though we intended it for the most underprivileged communities, the most affected schools could not afford the time and resources to organize and voice a problem they were clearly aware of. Sometimes what might seem like eye-catching, well-crafted dashboards and data visualizations could be confusing or irrelevant in practice for the citizens most affected. At times, the most affected citizens might not be the ones our technology should try to engage or mobilize first.

FOCO’s technology means we don’t need site visits to understand that resources are not being distributed as they should be, and yet we still visit schools and mobilize communities. Why? Because FOCO is centered around citizens. FOCO integrates and produces all the data necessary to expose corruption, including the perpetrators. If we had stopped at that, we would still have a shiny international example, but it would not reduce the power asymmetry or corruption. Alas, many civic tech stops there. Though inefficient and sub-optimal, mobilizing students and visiting schools is how collective action creates power, for power is not a monolith and it is created as soon as two or more people agree on something to advocate for. As a technology, FOCO was designed to reduce the cost of collective action so that people could create power that allows them to address their problem.

Civic tech can also unintentionally fracture processes both for the government agency and citizens. Reducing a 10-steps process to 3-steps is not always an improvement. Any public service delivery process—in all its glorious bureaucracy—has people-facing and desktop-based steps. While, yes, it is a welcome improvement that one can pay taxes and utilities online these days, generalizing this principle creates problems in other situations. For instance, education and healthcare delivery have many essential people-facing steps that are central to community-building. It is tempting to assume that if we reduce the “people” features of the process, corruption will have less room to grow. But when well-intentioned civic tech fractures a process to make it more seamless or efficient, it could be sabotaging the possibility of its own positive impact. It might be nice to avoid talking with school teachers or principals by having an app that allows you to directly report irregularities about the school… until you realize that these interactions were important levers to influence more transparency and demand accountability.

Understand power and community organizing

Civic technology, by design, should not work without citizen engagement. Moreover, if the civic tech does not have a community-building or collective action focus, it will most likely frustrate rather than empower the people who use it. This is why including a theory of power can propel us forward. Once we have mapped who has power and how our technology could help balance power asymmetries by facilitating collective action—which builds power by connecting people—then the technology’s design and implementation strategies will be more likely to reach its intended effect.

Only if we sustain collective action for long enough, can we increase the cost of political inaction. This switches our approach from focusing on what to show (i.e. through dashboards, data visualizations, etc.) to how to connect and sustain the connection that will be perceived as electoral risks. It reframes our vision from which inefficient processes must be optimized towards what existing good community practices could be amplified—or which related collective action costs are reduced. In short, we need to build for organizing, not viewing, and we must embed tools in existing community institutions, instead of inadvertently attempting to create new ones.

Conclusion

Corruption is not a pure technical problem. Civic tech that approaches it in that way will always be incomplete. When we think about civic technology, we must remember that we are talking about civic and software technologies at play. To design and implement something that addresses both we must start by assessing the existing, intricate fabric that make up civic technology and then think of software that could amplify the good while reducing the bad. At the end of the day, civic technology’s success should be measured by how much pro-social, offline activities it enables. Not the inverse.

David Riveros Garcia is a Researcher at the Ash Center’s Allen Lab for Democracy Renovation.

The views expressed in this article are those of the author(s) alone and do not necessarily represent the positions of the Ash Center or its affiliates.