Bridging the divide

New study shows how integrated institutions can lead diverse populations to cooperate in rebuilding countries.

One of the most pressing issues in world affairs today is state building: how countries can construct stable, inclusive governments in which a variety of religious and ethnic groups coexist.

Now a unique field experiment involving Muslim and Catholic students in Bosnia-Herzegovina suggests one avenue for building emerging states: The existence of integrated civic institutions such as schools, the study finds, helps foster greater collaboration across ethno-religious lines.

Such a result indicates that ethnic and religious identity need not be a decisive factor governing behavior even in conflict-torn regions, and that cooperation among different ethnic groups increases when those groups have greater social exposure to each other.

In the study, students from an integrated school were willing to make financial contributions to the public good that were as much as three times larger than the contributions made by students from segregated schools, among other effects. The results appear in a paper published this week in the journal Science, co-authored by MIT political scientist Fotini Christia.

"We actually see experimental evidence that integration works in promoting cooperation," says Christia, an assistant professor of political science. "It is a promising finding for multiethnic, post-conflict states, because it shows that integration can be a way forward."

Making free riders pay

Bosnia-Herzegovina, a state of about four million people in the former Yugoslavia, declared independence in 1992, but is still trying to overcome ethnic and religious tensions in the aftermath of the brutal Balkan Wars of the 1990s. The country has a mix of Muslim Bosniaks (who constitute about 40 percent of the population) along with large contingents of Serbs and Croats, who generally belong to the Eastern Orthodox and Catholic churches.

Christia and her co-author, Marcus Alexander of Stanford University, conducted the study in Mostar, a picturesque city of about 70,000 whose diverse population reflects its eventful past: At various times in history, Mostar has been part of the Roman, Ottoman and Austro-Hungarian empires.

Their field experiment was made possible by the judicially ordered integration of two of Mostar’s four high schools across ethnic lines. In other circumstances, comparing the effects of integrated and segregated schools can be problematic because of self-selection issues: People from families less inclined to cooperate with other religious or ethnic groups might choose to attend segregated schools. But in Mostar, the way the integration process took place meant that the students were, in effect, randomly assigned to either the integrated or segregated schools.

"It was this element that allowed us to examine the specific impact of integration," Christia says.

In the experiment, students participated in computer-based public-goods games, in which they were expected to contribute money. Christia and Alexander directed 26 sessions of experiments, each consisting of 20 rounds of the game, held among groups ranging from eight to 24 students.

The game’s structure allowed students to make voluntary payments to the public treasury, thus introducing the possibility of the "free-rider problem," in which some citizens benefit from the contributions of others.

The researchers found that students from segregated schools tended to make lower contributions than the other students when playing the game in ethnically mixed groups. Students from integrated schools made larger contributions, especially given one condition: the ability to impose sanctions punishing lower contributions. Those sanctions, in turn, were effective only in integrated settings.

"The effectiveness of sanctions, in inducing people to cooperate, can be very much dependent on the institutional context," Alexander says.

Overall, the finding implies a simple policy recommendation: The integration of schools and other public institutions has a clear impact on people’s willingness to contribute to the greater good in diverse societies.

Thus, while ethnic and religious tension is often seen as a challenge to effective state building, this issue can be mitigated. Or, as Christia and Alexander write in the paper, their research shows that "institutions may play an important role in creating environments in which diversity can be bridged... in order to promote well-ordered societies and better-functioning markets."

From Europe to Afghanistan

Scholars have found the study impressive. The paper "is distinguished by the quality of the field work," says Robert Bates, a professor in the Department of Government at Harvard University, and an expert in political development. In particular, he cites the fact that the researchers "employ an experimental design while also taking into account the nature of local institutions, history and culture."

Christia readily acknowledges some potential limitations of the study. Because the experiment was held in one locale, scholars may question how closely the results could be applied to other countries. Moreover, the institutions most in need of greater ethnic or religious integration -- schools, police or military forces, government bureaucracies -- may vary from state to state.

"Different countries have different conditions, and we may want to replicate this kind of social-science research in other post-conflict states," Christia notes. "That said, this finding still suggests that integration can work."

For her part, Christia is working on a book-length study of state building in Afghanistan, where she has conducted extensive research, and also evaluating whether other states, in the Middle East and North Africa, might provide opportunities for experiments similar to the one in Mostar.

Written by: Peter Dizikes