Increasingly in the developing world, when governments make local policy they are listening to local voices. But whose voices, exactly, get heard? Concerned that elites in Indonesia dominated decision making at the local level, Benjamin Olken, a development economist at the Massachusetts Institute of Technology, designed a field experiment to compare the effects of alternative democratic institutions. Would direct elections result in fairer outcomes, and happier citizens, than the current system in which a few representatives deliberate among themselves?
“Villages were deciding what kind of local public good they wanted to build,” says Olken. “They had a block grant and they could decide how they wanted to use the money, whether it should be to build a road, or a well, or an irrigation system,” or something else. In one of the first randomized field experiments of its kind, Olken designated the political process itself. He picked out 49 villages representing more than 100,000 people participating in the Indonesian Kecamatan Development Program (KDP), which is funded through a World Bank loan to finance small-scale infrastructure activities. Some of the villages continued to choose their preferred proposal at a meeting attended by a small group of village leaders—the usual KDP way. In the remaining villages, Olken set up direct election-based plebiscites in which every eligible citizen could vote.
“The key finding is that the plebiscite process resulted in dramatically higher levels of satisfaction and legitimacy of the program and of the proposal,” says Olken. Having had the opportunity to vote, the people in the study were more satisfied with the development program and judged the winning proposal fairer. They were more likely to agree that the project was “in accordance with the people’s aspirations,” that they would use the project, and that it would benefit them personally.
Interestingly, this was true despite the fact that villages chose exactly the same proposals through both political processes. They decided to build roads and bridges about 60 percent of the time and water and sanitation projects about 12 percent of the projects about 12 percent of the time, regardless of whether the decision was made at a village meeting or by a direct election with 20 times as many people participating.
That direct plebiscite did not change the ultimate decision is surprising. Still, it makes the increase in satisfaction all the more striking. “It’s some of the clearest evidence we have that the process can matter even if the outcomes don’t change,” says Olken. It shows that “direct participation can be a legitimizing force,” and when soliciting local input for community-driven development programs, “the details matter a lot. There’s a real difference between direct participation and indirect participation.” time, regardless of whether the decision was made at a village meeting or by a direct election with 20 times as many people participating.
Benjamin A. Olken, “Direct Democracy and Local Public Goods: Evidence from a Field Experiment in Indonesia,” American Political Science Review, 104, 2010.
Read more stories by Jessica Ruvinsky.