Smart African Politics: Candidates Debating Under a Tree - The N...

Transcription

1 FIXES Smart African Politics: Candidates Debating Under a Tree By Tina Rosenberg November 10, :30 am Fixes looks at solutions to social problems and why they work. Political debates are good even when they re bad. Even when candidates are cringe-worthy, they re cringe-worthy in public view. And voters learn about all the candidates, not just new ones. In the United States, for example, Hillary Clinton has been center stage in political life for 24 years. Donald Trump is the very definition of overexposed. Still, the debates tell us new things about them their positions, temperament, grace under pressure (or lack thereof), charisma and political skill. How much more could voters benefit from debates in countries where they know next to nothing about the candidates? Identity politics are everywhere. Instead of voting for candidates because they are competent or honest, or voting for those with congenial views on important issues, people often reflexively support the politician who shares their religion, ethnicity, race, tribe, clan or geography. But identity politics are strongest precisely where they can do the most damage. In countries divided by ethnic hatreds or fears, this kind of voting worsens the tensions that can lead to violence. It gives politicians no motive to be effective or upright, because they know that these qualities don t matter to voters. On the contrary, corruption and patronage can help them keep a majority group s support. 1 of 6 11/10/15 11:00 AM

2 We assume that this is unchangeable. To some degree, it surely is. But what if this kind of voting is in part a response to the fact that voters know nothing else about their candidates? They don t know candidates backgrounds, experience or education, much less their record or policy positions. So they vote on what they know: he or she is my candidate. Debates can help voters get the information they need. Matt Dippell, the global debate program adviser at the National Democratic Institute (he runs the website Debates International) said that at least 65 countries hold presidential or other national-level debates. Among them are many developing countries, including Nepal, Peru, Haiti, Maldives and Mongolia. And many are in Africa. Africa is a leader in this movement, he said. In the last 10 years there has been a remarkable wave of debates, and interest in holding more. But presidential debates aren t enough. Voters often do know something about presidential candidates. And they know something about local candidates, because these candidates are often neighbors. There s a big knowledge hole, however, in the middle: about candidates running for Parliament or Congress. Dippell said that at least 13 African countries have held or attempted some parliamentary or local debates, including Liberia, Nigeria and Malawi. Elsewhere, Haiti recently had a series of six parliamentary debates, with 24 candidates taking part. They were carried live on radio, TV and the Internet. Parliamentary debates are still rare, because holding them presents a far bigger challenge than presidential debates. Each constituency needs its own debates, and has very few resources to work with. So it s worth looking at the recently published results of an experiment in Sierra Leone s last elections. A nongovernmental group called Search for Common Ground held parliamentary debates in 14 competitive districts in the 2012 elections, and three researchers conducted a randomized control trial to see how much the debates changed the behavior of voters, candidates and elected officials. Search for Common Ground is a global organization that works on conflict prevention in 35 countries; it has been in Sierra Leone for 15 years. The researchers 2 of 6 11/10/15 11:00 AM

3 came from Stanford University s graduate school of business and two other American organizations, Innovations for Poverty Action, which managed the study, and the Abdul Latif Jameel Poverty Action Lab. Sierra Leone s presidential candidates debated in the 2007 election, but in 2012 the major parties boycotted and no real presidential debates were held. Also in 2007, Search worked with local groups and youth groups to organize some 15 live debates in parliamentary elections. When I.P.A. and J.P.A.L. approached us to study debates, we thought, this is great, said Ambrose James, who was country director for Search in Sierra Leone (he s now getting a master s degree at Harvard s Kennedy School of Government). We know candidates will come to the meetings, but we never knew about the impact. Search started by getting the go-ahead from the national electoral commission and leaders of the major parties. As is often the case, challengers were eager to debate, but incumbents who usually hold the advantage were not. In the end, all but one candidate, an incumbent, agreed to participate. (He not only disrupted the debate; he managed to get the participants and Search staff arrested on a false accusation that they were buying and selling voter registration cards. He won the election.) Most of the 45-minute debates were held outdoors, with the candidates sitting on chairs under a tree. A moderator from Search began by asking candidates to describe their background, education and interests: Do you have children? Do you play football? We started them with what they were familiar with, just to release tension and get them confident, said Saa Bandabla, the Search program manager for Sierra Leone, who headed the debate project. The substantive questions started with national policy on a new health care initiative and a proposed gender quota in public office, and candidates ideas for helping jobless young people. Next were two questions about local issues, tailored for each district. Candidates also had to explain how they would have spent (or had spent, if they were incumbents) the roughly $11,000 each member of Parliament received from the government for district projects. 3 of 6 11/10/15 11:00 AM

4 The sessions were actually question-and-answer, but at times turned into real debates. One candidate was explaining that he used his M.P. money to buy fuel to visit his constituencies, and others argued that the money was supposed to be for development, not transportation, said Bandabla. We allowed them to argue based on the issues we raised. Search videotaped the debates. But how could voters see the recordings? The villages that the researchers most wanted to reach had neither televisions nor electricity. So Search trucked screening equipment from village to village. When roads were impassable, villagers helped Search staff members carry the equipment in. The main screenings were at night, with the video projected onto the side of a building often the one where people would vote. Chiefs would send town criers to call everyone to come out and view the videos, said Bandabla. The atmosphere was usually festive, with people whooping and hollering for their candidate. The typical crowd was 100 to 200 people, said Rachel Glennerster, one of the researchers on the study and the executive director of J.P.A.L. The debates were all held in Krio, Sierra Leone s trading language. In most communities, the debate was shown twice first with pauses for translation into the local language, and a second time, sometimes without translation. In total they were screened in nearly 200 communities, reaching some 19,000 people. The debates increased voters knowledge of the candidates and issues, and appear to have changed the way people voted. Compared to voters in the same parliamentary district whose village did not see a debate, debate-watching voters were more likely to vote for the best candidate defined as the one the audience judged to have won. They were also 9 percentage points more likely, a huge difference, to vote for a candidate who shared their own political views. Researchers tested other media. To some voters who had not watched debates they showed only the candidates get to know me part of the video. They also interviewed voters who listened to a radio report that simply summarized the policy positions candidates took during the debate. In both cases, voters increased their knowledge but did not learn enough to change their voting habits. Radio is hugely accessible and democratic, said Mike Jobbins, the director of global affairs and 4 of 6 11/10/15 11:00 AM

5 partnerships at Search, in Washington. But people listen while they re cooking, while they re having conversations. It doesn t always capture the public like video. Candidates, too, changed their behavior because of the debates. They moved quickly to increase their campaigning in villages where the debate had been screened. They spent more time there, and gave these districts more gifts. Members of Parliament who debated spent 2.5 times more of their $11,000 on verifiable development projects for constituents than legislators who didn t debate. (Given that the money is paid directly into each legislator s personal bank account, it is remarkable that they spend anything on the district.) Once elected, the legislators also spent more time in villages that had watched the debates. Why did politicians pay more attention to debate villages? People were not necessarily going to vote along the patronage or ethnic lines that had historically characterized Sierra Leone s politics, said Jobbins. Politicians who had always counted on identity now knew they would be judged on more criteria. They had to campaign. Policy debate was new to village life. Candidates are appointed by their parties, said James. There is normally nothing around policy, nothing really substantial. They go around giving out bags of rice and T-shirts. The debates, he said, brought issues to the community level. This was an opportunity for people to actually listen to talk about gender, mining policy, transparency, what will you do with the money. Government is far away and abstract, he said. But this brings government down to people. And the people were not always pleased. If someone performed badly in the debate, said Badabla, people would say: In Parliament they re going to be talking about laws. How can it be that this person would go? Parliamentary debates could have a ripple effect. Katherine Casey, the Stanford researcher, found in her earlier work that voters are willing to vote across ethnic lines when one candidate is sufficiently superior but only when they have enough information. What if debates were taken to scale and made mandatory? said Casey. If 5 of 6 11/10/15 11:00 AM

6 candidates knew in advance there were going to be debates all over the place, I wonder if it would change the type of people who run for public office. Dippell points out that especially for countries emerging from violence, debates show a different way of solving problems. You have two candidates in a pitched battle, and you can show them on stage with a sense of mutual respect, while still disagreeing, he said. Often, he said, the moderators first question is whether all candidates pledge to respect the election results or, if they contest the results, to do so using only the established channels. The most important finding of the Sierra Leone study is that voting because a candidate shares a race, religion, ethnicity or patronage network may not be inevitable. People change their behavior in response to information, said Glennerster. It s telling us maybe lack of information is playing more of a role it s not just that everyone is being bought and doesn t care about policy. It s that they can t vote on policy. Join Fixes on Facebook and follow updates on twitter.com/nytimesfixes. To receive alerts for Fixes columns, sign up here. Tina Rosenberg won a Pulitzer Prize for her book The Haunted Land: Facing Europe s Ghosts After Communism. She is a former editorial writer for The Times and the author, most recently, of Join the Club: How Peer Pressure Can Transform the World and the World War II spy story e-book D for Deception. She is a co-founder of the Solutions Journalism Network, which supports rigorous reporting about responses to social problems The New York Times Company 6 of 6 11/10/15 11:00 AM