Designing Democratic Participation

It may take well-designed policies to insure fair representation to the poor, women and minorities.

Why doesn’t democracy work better? Ideally, it should be an equalizing force in society: once universal suffrage is achieved and elections aren’t fixed, citizens should be able to use their vote to “hire” quality legislators who represent their interests.

Yet surveys show that particularly in low-income democracies, citizens consider political parties and elected officials highly corrupt. In slum areas, the poor majority of residents should be able to use their vote to demand quality public services, yet elite groups manage to manipulate the vote and channel resources elsewhere, leaving public services in deplorable conditions. Women should have a representative voice in governments, yet nearly a century after enfranchisement, female presence in parliaments, even of such advanced Western democracies as the UK and US, remain low: 22.5% and 18%, respectively.

One approach to protecting the rights of the poor and ethnic minorities is the decentralization of power. The village council knows better than parliament what the local needs are; our work in Indonesia has suggested that local leaders may outperform central governments as fair and efficient distributors of welfare programs. Furthermore, even if the national assembly is ruled by the ethnic majority, local councils will represent local populations. Yet decentralization carries its own risks of low-level corruption and the unfair allocation of resources. Our work in India suggests that locally-based goods distribution may favor those who live near the elected representative or share his group identity. Further research shows that decentralization makes local councilors themselves more likely to be selected for welfare payouts and their villages more likely to receive government projects. All in all, democracy has some way to go to fulfill its definition as “rule by the people,” particularly in low-income settings.

Can well-designed policies act as add-ons to make the system more fair and inclusive? And what effect does political participation beyond voting – such as attending protests or standing for office oneself – have on policy and culture?

Informing voters as a means to increase quality representation
Country: India  |  Researchers: Rohini Pande, Michael Walton, Abhijit Banerjee

EPoD, Informing Voters as a Means to Increase Gender ParticipationIn a field experiment, we showed that providing Delhi slum residents with report cards on their legislators’ spending decisions and performance – as well as financial and criminal records of top candidates – increases accountability. For the best-performing incumbents, vote share increased by 7 percentage points. The evident sophistication in interpreting information weighs against the view that the poor and illiterate cannot be effective voters.

Policy Partner: Satark Nagrik Sangathan (SNS), Delhi
Research Partner
: J-PAL South Asia at IFMR, Chennai, India
: USAID-DIV, National Science Foundation, and the International Initiative for Impact Evaluation (3ie)
Exploring the wide effects of gender quotas
Country: India  |  Researchers: Rohini Pande, Lori Beaman, Raghabendra Chattopadhyay, Esther Duflo, Petia Topalova

EPoD, Exploring the wide effects of gender quotasWe have based several studies on a natural experiment created by an Indian law mandating female participation on village councils. These quotas create a propensity for positions of power that outlast the mandate, erode men’s biases, and increase parents’ career aspirations for their daughters – as well as girls’ for themselves. Further, the effect closed the gender gap in adolescent educational attainment in treatment villages.

Researchers: Rohini Pande, Lori Beaman (Northwestern), Raghabendra Chattopadhyay (IIM Calcutta), Esther Duflo (MIT), Petia Topalova (HBS)
Research Partner
: J-PAL South Asia
: Ash Center Harvard, Nike Foundation, MIT, YCIAS Yale, National Institute of Child Health and Human Development, and UNICEF

Quantifying the effects of political protests
Country: United States  |  Researchers: David Yanagizawa-Drott, Andreas Madestam, Daniel Shoag, Stan Veuger

Until now it has been unclear whether protests actually cause change or simply express change that is already underway. But a new study uses rainfall-related variation of turnout at Tea Party demonstrations in the US on Tax Day 2009 to show that the presence of one protester at a rally caused well above one increased vote. Attendance also affected contributions to the Tea Party cause and the decisions of Democrat incumbents to retire.

Experimenting with participant-driven evaluation
Country: Kenya  |  Researchers: Ryan Sheely

EPoD affiliate Ryan Sheely is educating community members to involve them in research as a means to improve data and hold politicians and civil society organizations accountable for project performance. Preliminary results suggest that a broad cross-section of study subjects are able to grasp core research concepts and methodologies, and that participation can empower community members and change the way they interact with politicians and NGOs.

Policy Partner: The SAFI Project, Laikipia Region, Kenya
Funding: William F. Milton Fund (Harvard), The Weatherhead Center for International and Area Studies (Harvard), Ash Center for Democratic Governance and Innovation (Harvard)