Democracy Lab

Avoiding War Number Two in Liberia

A war-torn country is not a broken country. How Liberia pulled off its 2005 election.

Note: This article is an abridged version of a longer historical case study produced by Innovations for Successful Societies, a research program at Princeton University.

In August 2003, a peace agreement ended the civil war that had ravaged Liberia since 1989. In a country of roughly three million people, an estimated 250,000 had died in the conflict, and hundreds of thousands were displaced. The peace accord dissolved the government of Charles Taylor, who went into exile in Nigeria. The accord also cancelled the presidential elections that were slated for the fall of 2003 and named Charles Gyude Bryant as head of an interim government, which was responsible for implementing free, fair, and comprehensive elections in the recovering country.

To monitor the implementation of the peace agreement, the United Nations deployed 15,000 troops and 1,100 police -- at the time the largest peacekeeping mission in the world. Hundreds of civilian U.N. employees also assisted in aspects of post-war reconstruction. Taken together, the military and civilian operations constituted the United Nations Mission in Liberia (UNMIL).

The peace agreement banned officials of the interim government from running for office, dissolved the existing elections commission, and required Gyude Bryant to appoint seven new election commissioners to be confirmed by the interim Assembly. In an effort to create a neutral elections commission, the peace agreement stipulated that the commissioners be recommended by civil society groups and have either civil society or justice backgrounds. Gyude Bryant nominated Frances Johnson-Morris to head the new election commission. She had served as Chief Justice of the Supreme Court in 1997 during a lull in the civil war, and later headed an NGO dedicated to combating human-rights abuses. The Assembly approved her appointment in April 2004 -- 18 months before the crucial national elections. Johnson-Morris had built wide respect as a neutral broker during her tenure as chief justice, engendering trust in her ability to work in the national interest.

The interim government sought to draft a new and clear electoral law to govern the upcoming elections. Legislators debated between using a proportional representation (PR) system or a majoritarian one. The last general elections in 1997 used an "open list" PR system, meaning that candidates were not pre-assigned to a constituency or county; the consequence was that candidates looked to curry favor with political superiors who could dictate their inclusion on the list, rather than constituents. After hearing from commissioners and election experts, the legislators decided on a majoritarian system in which candidates for all offices ran in a single-round election, hoping it would create stronger ties between individual candidates and voters.

The legislature passed the electoral reform law in December 2004. It also abolished a requirement that presidential candidates live in Liberia for at least 10 years, thereby encouraging newly-returning members of the diaspora to compete for public office. "Everyone saw this as an opportunity to participate," Johnson-Morris recalled. "Some people felt that they had been marginalized and excluded for so long. Everyone was eager to go to the process." The electoral commission received considerable assistance from international groups like the United Nations Electoral Assistance Division (UNEAD), the United Nations Development Programme (UNDP), and the International Foundation for Electoral Systems (IFES), which provided the commission with technical and logistical support, and contributed more than three-quarters of the $18.9 million election budget.

Johnson-Morris and the election commission had to work quickly to prepare the country for elections. Liberians were on edge. Years of war had sown deep mistrust between ethnic factions. Partisanship had undermined the credibility of previous commissions. Even if the elections commission acted without bias, Liberians were uncertain whether opposing factions would respect the integrity of the vote. As a result, Johnson-Morris set a high priority on building the public's trust in the commission and took steps to ensure the elections were as inclusive as possible.

She and her seven-member team faced a number of hurdles. First, finding qualified election staff would be difficult. Much of Liberia's educated class had fled the country, leaving a relatively small pool of qualified individuals. Second, much of Liberia's transportation and communication infrastructure had been damaged or neglected during the war, and large parts of the country were inaccessible by road. The elections commission needed to register and educate people about the elections; contacting them would be problematic. Moreover, voters had to be able to reach polling places. Adding to these difficulties, mobile phone networks covered only half of the country, and few places outside Monrovia, the capital, had access to the Internet. Third, the commission had to manage a diverse collection of political parties and hundreds of candidates. Most had no experience in campaigning, and political parties were often formed around personalities rather than interests or issues.   

To tackle these challenges, the commissioners met every Tuesday to propose, debate, and vote on new rules to guide the elections. They realized that in order to win public trust, they needed qualified and apolitical staff at their headquarters. The commission advertised for positions, reviewed applications, and shortlisted candidates. They interviewed finalists, who completed a written assessment of their ability to read and write in English. The screening committee disqualified any candidates who expressed partisan views. Many of the new hires had backgrounds in civil society, but few had practical experience in elections management. The commission's international partners trained the headquarters staff on operational issues including voter registration, boundary delimitation, and polling procedures. 

To oversee elections on the county level, the commission appointed magistrates, who were required to hold bachelor's degrees and work in the county in which they lived. United Nations civil service workers assisted the magistrates in hiring a total of 19,000 poll workers. Given the shortage of educated workers, particularly in rural areas, the election commission set few requirements for poll workers. Magistrates gave preference to teachers and other educated professionals, but considered candidates who displayed a basic ability to read and write.

In registering political parties, the commission required new parties to submit lists of at least 500 voting members from 12 of Liberia's 15 counties, rather than from six counties as previously required. Although this provision appeared more onerous, the commission purposefully did not scrutinize the list of voting members in an effort to promote inclusiveness; any move to keep a significant party out of the election could result in disputes and delay the elections process.

The commissioners sought to enhance communications with the parties by creating a consultative group, the Inter-Party Consultative Committee (IPCC). Modeled after a successful experiment in Ghana, the committee was designed with input from both the election commission and the political parties. They agreed that each party would send two representatives to a monthly committee meeting, at which any issues could be raised. Commission staffers would then issue a press release to document resolutions. "The intent behind the Inter-Party Consultative Committee was to defuse tensions and minimize suspicions and misconceptions between the elections commission and the political parties," Johnson-Morris explained. "We wanted to work as partners, not antagonists."

Through the IPCC, the parties agreed on a code of conduct, and commission staffers used the meetings to familiarize the parties with specific aspects of election-day procedures. The gatherings also provided an opportunity for the electoral commissioners to assert their authority. They warned party leaders that they would revoke the accreditation of parties and candidates involved in election-related violence or harassment.

To encourage broad voter participation, the commission sought to register as many eligible voters as possible. Liberia's census figures -- from 1984 -- were too old to be useful in drawing electoral boundaries; the commissioners decided that a voter registration drive would produce the data they needed as a proxy for a full census. They recruited registration officers, established voter sign-up centers, and trained the officers in registration procedures.

During a 30-day period between April and May 2005, 1.3 million Liberians -- 90 percent of eligible citizens -- registered as voters. Village elders vouched for the age and residency of people who lacked identifying documents. Commission staff recorded people's names and other relevant information, and took two photos. One photo accompanied the registrant's name on a voter roll, and the other was put on an identification card given to the registrant on the spot. Registration staff then sent the voter rolls to the commission's headquarters, where staff entered the names and photos into a computer system. Complications, such as cases of multiple registrations, were investigated by the district magistrates.  

While voter registration efforts were largely successful, a major challenge arose from displaced persons. Many Liberians who were displaced by the war and living in camps refused to register unless UNMIL moved them back to their original homes. They feared that after the elections, the new government would give little attention to their plight. But returning all the displaced people to their homes would be impossible before the October elections; many of their villages had been severely damaged and lacked essential infrastructure. The elections commissioners traveled to the camps and urged people to sign up to vote, but many balked. Johnson-Morris faced the prospect of tens of thousands of eligible voters refusing to participate in the elections. 

The commissioners consulted with the transitional government and decided to allow people in camps to register in the counties to which they would be repatriated. For their votes to count in the House and Senate elections, the displaced people had to cast their ballots in their home counties. Displaced people could vote for the president and vice president regardless of whether they were still in the camps or back home at the time of the elections. This arrangement still failed to satisfy many, and the issue continued to simmer for months. With the elections nearing, Johnson-Morris turned to someone she knew had a strong rapport with the displaced communities: Brigadier General Luka Nyeh Yusuf, a Nigerian with the UNMIL peacekeeping force responsible for ensuring the safety of the camps. He held multiple meetings with senior leaders in the various camps to discuss issues of concern, and after months of persuasion, the leaders agreed to encourage the registration effort. In the end, about 61,000 of the estimated 140,000 people living in formal camps in Liberia signed up to vote.

In addition to voter registration, the commission tackled the difficult issue of electoral boundaries. John Langley, the commission's senior policy adviser, indicated the process was a collaborative effort by the agency and local communities. "We could not go and impose on citizens," Langley said. Commission officials held town-hall style meetings in local communities, where they presented rough outlines of the new boundaries and solicited ideas and questions. When commission staff returned to headquarters, they worked with international experts from UNMIL and IFES to nail down the boundaries for each district, taking into account social and cultural factors noted during the town hall gatherings. An IFES official explained the collaborative approach: "The village chiefs would come and say, 'No, that area there, the people over there will not feel satisfied if they are joined with these people.'"

On election day, the commission worked hard to prevent voter fraud and suspicions of fraud by monitoring polling staff and mandating adherence to electoral procedures. To assist illiterate voters, the commission printed candidates' photos and party symbols next to their names on the ballot. Still, many voters did not know how to mark their ballots. The commission therefore allowed polling station officers to instruct voters on the proper way to mark their ballots rather than risk widespread ballot invalidation. When the polls closed, party agents signed the official tally sheets to certify they had witnessed the counting of each ballot. Poll workers placed one copy of the tally sheet into a transparent ballot box along with the ballots, and sealed the box. Each party agent received a copy of the signed count. Poll workers then transported the box to the county elections office. The commission did not recount the ballots at the county elections office, nor did it undertake a parallel vote tabulation.

A cadre of international and domestic watchers helped ensure proper conduct of polling staff. Commissioners had visited their assigned counties in the weeks before the elections, observing training sessions and meeting with the magistrates to review preparations. Their presence delivered a clear message to poll workers that headquarters was carefully watching their actions. International forces provided election security, as the Liberia National Police were severely understaffed and under-resourced. Peacekeepers were stationed at every polling place, while mobile patrols kept watch over access points to and from the voting centers. 

Thanks in part to these pre-emptive measures, the October elections unfolded relatively peacefully. However, tensions flared in November, when presidential candidate George Weah lost a run-off vote to his chief opponent, Ellen Johnson-Sirleaf. After the commission announced the run-off results, thousands of Weah supporters marched on the U.S. Embassy, under the supposition that the United States had rigged the election for Johnson-Sirleaf -- an idea espoused by the chairman of Weah's party. Protesters threw stones and police responded with tear gas. Weah appealed for calm while he challenged the result at the elections commission, which found insufficient evidence of fraud. He then appealed to the Supreme Court, but dropped the matter after the African Union pressed him to concede for the sake of peace.

Despite this controversy, the elections commission was satisfied with the conduct of the elections. Samuel Cole, the director of civic and voter education, noted proudly, "Today, people can look at the 2005 elections in Liberia as a role model in Africa." Although there were scattered incidents of violence before and after the voting, several factors contributed to the relative peace. The heavy presence of armed UNMIL peacekeepers prevented local disputes from worsening and gave voters confidence that they would be safe. The success of the elections also reflected the commission's care in staff selection, its emphasis on fairness for parties and candidates, its openness, and its willingness to relax certain requirements to encourage broad participation by Liberians.

Speaking in 2011, Frances Johnson-Morris recalled voters' enthusiasm for the 2005 elections. "Those were exciting times, because Liberia had been at war for nearly 14 years, and to have an opportunity to elect leaders that would take them out of the mode of belligerence and violence, that was very good," she said. She stressed that the commission's consultative approach earned the trust of voters and political parties. "When you involve people in decision making, you uphold your transparency and your credibility. I think we did that very well." 

Photo by Chris Hondros/Getty Images

Democracy Lab

Changing the Code on Corruption

How a Brazilian government commission tried to fight back against sleaze.

Note: This article is an abridged version of a longer historical case study produced by Innovations for Successful Societies, a research program at Princeton University.

In February 2001, Brazilian newspapers noted something different in Rio de Janeiro's annual Carnival celebration: The public officials who traditionally populated corporate boxes were no longer present. In the past, these high-profile officials received VIP treatment, with private beer companies paying for their airfare, meals, and reserved seating.

The difference hinged on Brazil's year-old Public Ethics Commission. A few months earlier, the commission had published a rule that clarified what gifts, if any, senior civil servants could accept, deeming numerous perks unacceptable because they could bias politically-sensitive decisions.

The rule irked powerful individuals in the private and public sectors. The media exposed three officials who flouted the rule by accepting box seats owned by Brahma beer. The commission reviewed their cases, requested explanations, and issued warnings that discouraged future violations. After the officials apologized, they promised, in writing, to obey the rule in the future. The rule and its enforcement earned public support, with one newspaper survey indicating a 98% approval rating. Senior officials now had to think twice before accepting such perks.

Changing ingrained political behavior was no easy task. After Brazil emerged from decades of dictatorship in 1985, pervasive political scandals detracted from the new democratic emphasis on accountability. Politica de favores, or the politics of favors, rendered merit-based hiring, procurement, and privatization uncompetitive and inefficient. Officials accepted bribes thinly disguised as gifts and gave jobs to relatives as a matter of familial allegiance. Without a formal system of government ethics, doing the "right thing" often was considered foolish, impolite, or offensive. Frequent scandals involving nepotism and influence-peddling weakened public trust in government officials and embarrassed successive administrations. Although the constitution of 1988 set stringent rules and harsh penalties regulating civil servants' behavior, they were rarely enforced.

In the mid-1990s, the administration of President Fernando Henrique Cardoso generated a clear impetus for change with a modernization, privatization, and government-reorganization drive. The administration was forced to take action after internal scandals exposed unseemly behaviour by senior government officials and cast public doubt upon the administration's initiatives. "We felt that Brazilian society was changing with modernization and was no longer willing to accept such behaviors," said Cardoso's chief of staff, Pedro Parente.

Cardoso appointed Luiz Carlos Bresser-Pereira, a well-known Brazilian economist, to spearhead broad civil service reforms and amend the human-resource rigidities of the 1988 constitution. In May 1995, Bresser-Pereira formed the State Council of Reform to devise options for public-sector reorganization. João Geraldo Piquet Carneiro, a former minister in charge of a large-scale de-bureaucratization effort in the 1980s, was one of the earliest council members. Piquet approached like-minded members on the council and quickly brought the issue of officials' conduct onto the agenda. He then created a small team within the council to work on a code of conduct.

The team faced a daunting task. They sought to create a code that could reach and influence a large number of public officials at a reasonable cost. (Brazil's civil service consisted of nearly 600,000 people.) Second, the team members realized they would have to set benchmarks for situations in which officials' personal interests clashed with their public roles. Senior civil servants, accustomed to significant freedom in decision-making, were likely to offer the most resistance. Third, the code needed to specify reasonable parameters of behavior when the issue involved shades of gray. For example, when it came to the acceptance of gifts, at what monetary amount or under which circumstances was a gift considered improper? How long should a public official reasonably wait after leaving public office to accept a private-sector post? Finally, the code required an enforcement mechanism.

The issue fell to the back burner, but during a television interview in early 1999, a journalist asked President Cardoso about the code of conduct, sparking sudden media and public interest and bringing the code back into the limelight. Cardoso issued a presidential decree forming a new advisory body, the Public Ethics Commission, to work on ethics-related issues and implement the code of conduct, with Piquet as its first president.

All commission members were unsalaried volunteers appointed by Cardoso, and were prominent Brazilians who enjoyed reputations for integrity and efficiency. Few had government experience: Piquet believed that a team with no direct ties to the Cardoso administration would earn credibility for being independent of partisan interests. With a limited budget, they decided to target only senior government officials: presidential appointees, ministers within the Civil Cabinet, managers, and directors of public organs and companies, with the hope that senior officials could set an example for the government and potentially improve public opinion of civil servants.

Studying other countries' codes of conduct, the final code set guidelines on receiving perks, reporting assets, investing in assets while in office, maintaining shareholder stakes in companies while in office, leaving public office for private positions, and publicizing conflicts of interest. The team also began to work on an enforcement strategy. Piquet organized a seminar for the drafting team, officials from the civil service, and international organizations to share ideas and models that could work well for Brazil. "Given the traditions and realities of the Brazilian political system, we wanted to see what could be adjusted and innovated for Brazil from international experiences," explained commission member Lourdes Sola.

In August 2000, another presidential decree established the code of conduct. Piquet's team now faced six tasks: developing institutional capacity, communicating and publicizing the information among civil servants, monitoring whether individuals were following the code, establishing procedures to deal with violations, facilitating understanding and clarity of the code, and developing a system of ethics that would extend beyond high public officials.

The six commission members quickly realized the need for staff support; they all lived in different parts of the country, and their monthly meetings were not sufficient to enforce a code intended for all senior civil servants. They selected Mauro Sérgio Bogéa Soares, a former tax auditor and public servant known for his attention to detail, as their secretary. "Any successes we had ... were mostly due to him and to Piquet, and a very motivated board," said commission member Roberto Teixeira da Costa. Soares meticulously conducted an informal count of officials who were covered by the code, and within two months of his appointment, identified 787 officials. He set about publicizing the commission's work and inundating ministers and officials with copies of the code. He also met with ministries to explain the code, portraying it as a source of guidance rather than as a regulator of behavior.

The commission then devised ways to enforce the code. Civil servants who were already in office were to sign and submit the code within 30 days; Incoming hires had to submit the signed code within 10 days of their appointment. Failure to comply could result in the commission recommending their removal. A Declaration of Confidential Information required officials to submit details of their finances with the signed code. Senior government officials tended to bring "private baggage" when they entered office -- many were former chief executives of companies or public banks. The commission would work with officials to untangle their private and public interests if conflicts arose. The staff studied the declarations meticulously and became adept at detecting conflicts of interest -- and also uncovered attempts to subvert the process.

Informing officials about the code and the declaration was a difficult task. Many accused the commissioners of mistrusting the president's judgement, as many senior officials were presidential appointees. But President Cardoso and Chief of Staff Parente stood firm, encouraging all ministers to comply. Piquet and Soares would also meet with non-cooperative ministers, explaining the consequences of failing to comply. "It was a very difficult public position [for senior ministers] to hold, to say that you did not want to behave in a proper manner," Sola said.

Reports of infractions could come to the commission either through a complaint from an official's colleague or, more often, through the media. As soon as the commissioners discovered a violation, they collected as much information as possible on the situation, and would then issue a warning to the official involved and request an explanation. If the official's explanation was satisfactory, the case was closed; if not, the commission took disciplinary action, sometimes making an official donate to charity or reimburse public coffers. If the official ignored the initial notice, the commission issued a more stringent warning and informed the president's chief of staff. If the official ignored the commission a third time, the commission advised the chief of staff and the president to dismiss the offender; presidential support was thus vital to the model's success.

From the start, the commission faced a dilemma. If it was too strict, it would alienate civil servants, but if it was too lenient, it would lose purpose and efficacy. It had to balance the roles of watchdog and collaborator. Piquet stressed the importance of building respect through a flexible, low-key approach of consultations and conciliation. "It was important for the authorities to perceive that we were not witch hunters, [and] that they would have the support of the commission if something happened," he said.

Piquet arranged several highly publicized demonstrations of how individuals complied with the code. One such case involved the president of the Central Bank of Brazil, Arminio Fraga: His position was fraught with potential conflicts of interest because he had advance knowledge of economic indicators like inflation and exchange rates. Fraga volunteered to transfer management of his financial assets to an independent third party. This public display of a high-profile financial management agreement set a standard for how economic decision-makers could separate their public and private interests; it was also the first time that a Brazilian minister discussed the management of his own assets.

Piquet looked to Brazilian culture to illustrate why a conciliatory system enforced by social pressure worked better than rigid legal enforcement. He referred to Rio's famous Samba dance halls, where rowdy behavior evoked warnings from both hall officials and dancers. If the misdemeanor persisted, fellow dancers ejected the offender. This shared enforcement system effectively suppressed misbehaviour better than rigid laws; it was a way of changing behavior by taking a middle path, acting as a mediator rather than an enforcer.

Much of the commission's enforcement system hinged on trust: The commissioners had to demonstrate a degree of trust in officials' compliance. Piquet felt that this trust-based system generated mutual respect, but critics doubted its stability and said it created room for duplicity. The team largely overcame this challenge with the support of the president and the chief of staff, as well as the media's watchdog role as an indirect enforcer of the code.

The Public Ethics Commission set a clear precedent in the Brazilian political system, eliminating a gray area in officials' duties and holding those in the upper echelons of the civil service more accountable to the public. During the first three months after the ethics code's adoption, over 40% of the 787 officials failed to submit their financial declarations. Instead of issuing stern warnings, the commission sent reminders to the defaulters and arranged face-to-face interviews with the few individuals who repeatedly ignored the code. Within six months, the level of missing financial declarations fell to a rate below 1%.

The commission spread ethical standards by encouraging decentralization, routinely meeting with heads of government agencies to encourage them to set up their own internal control mechanisms, codes of conduct, and ethical panels. Ethical behavior became the norm as opposed to the exception among the upper echelons of the civil service: By 2010, commission members and officials acknowledged that gifts and perks were universally frowned upon in Brazilian political culture. Fernando Neves da Silva, who served as the commission's president after Piquet, said, "The commission can really be an entity that bothers politicians, like a mosquito buzzing in your ear."          

While enthusiastic about the commission's overall impact and the endurance of its early successes, several former commissioners expressed disappointment regarding its operation under Cardoso's successor, President Luiz Inácio Lula da Silva. At the beginning of President Lula's tenure, the commission was still operating strongly under Piquet's leadership. After Piquet left in 2004, however, it lost momentum. From 2006 to 2008, the commission operated with only three members; Lula claimed that he could not find people willing to take the unsalaried positions. Additionally, there were cases where the president was not supportive of commission decisions, causing commissioners to resign in protest. The commission regained some momentum in 2009, but the divergence in its performance between the Cardoso and Lula administrations highlighted the critical importance of a strong working relationship between the president, the chief of staff, and the commission.

Piquet offered a modest but positive assessment of the commission's impact: "We kept worrying about very ordinary things, which doesn't compare to some of the large scandals in Brazil," he said. "But if you have a government function and don't regulate the gray areas, your area of contamination becomes a lot larger. Nobody had done it before ... but we did a little bit. That's certainly better than nothing. We created a background reference that is still important today."

Photo by PEDRO LADEIRA/AFP/GettyImages