RECLAIMING THE CITY:
POLICE REFORM IN MEXICO CITY, 2002-2008

SYNOPSIS
When Andrés Manuel López Obrador became mayor of Mexico City at the end of 2000, a massive crime wave was sweeping the national capital. From 1995 to 1998, the city's overall crime rate had nearly tripled. Aware that taking back the streets from criminals would require a new approach, López Obrador brought in an experienced political leader, Marcelo Ebrard Casaubón, to head the Secretariat of Public Security of the Federal District. Together they introduced new systems that could document, map, and analyze crime and lead to more-efficient allocation of police resources and better preventive policing strategies. Ebrard also engaged a team to create Community Protection Units, improve police-citizen relationships, professionalize the police, and build a neighborhood police program. Despite abrupt leadership transitions at the public security secretariat, a decade later Mexico City had greater capacity to combat crime and greater political control over high-crime areas of the city, thereby laying the foundation for additional public security reforms.

Rachel Jackson drafted this case study based on interviews conducted by Ben Naimark-Rowse in Mexico City, Mexico, in February 2013 and by Jackson in November 2013. This ISS case study was made possible by support and collaboration from the Monterrey Institute of Technology and Higher Education. Case published June 2014.


INTRODUCTION
    Even in a country that ranked first among members of the Organisation for Economic Co-operation and Development for violent robberies, in the late 1990s Mexico City stood out. Daily life in Mexico's capital city carried the ever-increasing threats of kidnappings, assaults, armed robberies, and murders. 
    Although insecurity had plagued Mexico City for decades, the 1990s brought rising crime rates across the board. From 1990 to 1996, reported rates of robbery, property damage, fraud, and extortion in Mexico City more than doubled from 1,059.0 incidents per 100,000 inhabitants to 2,434.3, and the percentage of robberies involving violence jumped.1 And the trend worsened in the latter half of the decade: from 1995 to 1998, Mexico City's overall crime rate nearly tripled. By the late 1990s and early 2000s, Mexico City's residents were expressing higher levels of perceived insecurity than in any other city in the country.2
    In December 2000, the city's voters elected Andrés Manuel López Obrador as mayor. A member of the left-wing Party of the Democratic Revolution (Partido de la Revolución Democrática, or PRD), López Obrador made taking control of the Mexico City Police and tackling crime priorities for his administration. He publicly acknowledged that the city's deteriorating public security could undermine the government's control of the city itself and that dealing with the situation was crucial to the future of his minority political party.3
    "López Obrador caught on to the idea that in order to have greater political control of the city as well as to have a political future for the left, he needed to really engage on security issues," Colegio de México professor and policing expert Arturo Alvarado said.
    Initial progress on security reforms stagnated, however, when the head of the Secretariat for Public Security of the Federal District (Secretaria de Seguridad Pública del Distrito Federal, or SSPDF), Leonel Godoy Rangel, resigned from the administration to join the state government of Michoacán. 
    In February 2002, López Obrador put his reform agenda back on track when he picked Marcelo Ebrard Casaubón, a respected former political rival, to head the public security secretariat for the district. At the time, Ebrard was a member of Mexico's lower house of Congress, where he had led investigations into government corruption. Because he was affiliated at the time with a newly formed smaller political party rather than one of the three large parties that jockeyed for control of Mexico's government-the PRD, the president's National Action Party (Partido Acción Nacional, or PAN), and the former ruling Institutional Revolutionary Party (Partido Revolucionario Institucional, or PRI)-some observers viewed his appointment as a compromise between the Mexico City administration and the federal government.4
    Ebrard had gained experience in managing public security issues in the city. In the early 1980s, he joined the secretariat of planning and budget and by 1992 had risen to secretary of government. Five years later, he won a seat in the national legislature. In 2000, Ebrard formed a new political party and ran for head of government (mayor) of Mexico City, but he later withdrew and endorsed the leading candidate, López Obrador.
    "[López Obrador] needed someone with political leadership [skills]," Ebrard said of his later appointment as secretary of public security. Those skills would prove indispensable in the task that lay ahead.
    Reforming the police and dealing with Mexico City's rising crime rates would present Ebrard with immense challenges that had defeated other would-be reformers. In 1998, Mexico City chief of police Rodolfo Debernardi had resigned after only a year in office, stating that the person capable of dealing with the city's crime problem had yet to be born.5

THE CHALLENGE
    López Obrador's law enforcement views played a major part in his administration manifesto, titled "General Plan for Development of Mexico City 2001-2006." The report identified a lack of effective preventive policing and low levels of citizen participation in crime prevention as major challenges. It also highlighted the low technical capacity of the police and the need for advanced monitoring methods to combat crime. 
    The responsibility for dealing with Mexico City's burgeoning crime problem fell largely to the local police. Even with nearly 90,000 members-34,000 preventive police; 40,000 auxiliary police who guarded official buildings, the airport, and other specific locations; and 15,000 banking police-this was no small task. In 2000, the city-politically and 

administratively constituted as the Federal District and separated from the country's 31 federal states-was the home of more than 8.6 million inhabitants within 1,470 square kilometers (571 square miles), with an additional 10 million living in its greater metropolitan area. The ratio of preventive police per 100,000 population was lower than it was in the city of New York, and the preventive police service covered a geographic area nearly twice as large.6
    Under the aegis of the public security secretariat, Mexico City's preventive police patrolled neighborhoods, responded to emergencies, and dealt with crimes in progress. However, different parts of the city government handled other policing functions. Under the separate administration of the attorney general of the Federal District, the judicial police-roughly 3,500 officers-had the power to conduct investigations and build legal cases against individuals and organizations. Cases of drug trafficking and public corruption were in the hands of the country's federal police services.
    The preventive police were ill equipped to address the growing insecurity effectively. Disorganization, poor training, lack of performance-based advancement, and minimal accountability mechanisms meant there were few incentives for police officers to perform effectively and honestly. Although recruits earned around 4,000 pesos (US$415 in 2000) a month (above Mexico's moderate urban poverty line of 827 pesos (US$90) per month in 20027), officers frequently worked second jobs or supplemented their incomes through bribes to support their families. 
    "[The low salary] makes the police susceptible to crime because people are beholden to who pays them the best," said Joel Ortega, who advised López Obrador on public security and later became secretary of public security. A scholarly analysis of the Mexico City police published in 2000 found that recruits most commonly entered the police "for the money" despite low official salaries.8
    Widespread perceptions of police corruption contributed to a tense and often antagonistic relationship between Mexico City residents and the police, particularly in poorer neighborhoods. A 1999 survey by the Mexico chapter of Transparency International found that 90% of Mexico City respondents had "little to no trust" in the Mexico City police.9 The high level of distrust discouraged high-quality candidates from joining the preventive police and sharply reduced the level of public cooperation required for effective community policing.
    Furthermore, the police had neither the systems for collecting basic data about crimes, such as the sociodemographic characteristics of victims and alleged criminals,10 nor the capacity to analyze the information. The deficiency was partially a technological shortfall, because many police stations had little or no computer capacity. But the broader problem was administrative: the police lacked a central point for collecting and analyzing the city's daily crime data.
    Ebrard also faced likely resistance to reforms in the public security secretariat itself. In the late 1990s, when Mexico City's leaders attempted to implement mandatory lie detector tests and restructuring deployments that would create accountability and would curb corruption, the police went on strike and crime rates immediately jumped. 
    A final challenge involved possible opposition from those Ebrard was trying to help; past efforts to increase policing in some of Mexico City's most dangerous neighborhoods had met resistance from the communities themselves.11

FRAMING A RESPONSE
    To implement effective police reform in Mexico City, Ebrard needed a comprehensive 

security strategy that had strong backing both within the López Obrador administration and from Mexico City's powerful business interests. Ebrard also had to recruit a high-performing and reliable team that could help him mold the preventive police into a more effective law enforcement organization.
    In early 2002, when Ebrard arrived at the Secretariat of Public Security and took stock of his challenges, he decided that Mexico needed an entirely different way of thinking about policing and public security. "I noticed that for 20 years, everyone had repeated the same idea," he said, recalling early conversations with the police leadership. "I said, 'Well, something is not working very well because in the past 20 years, you are saying to me that we are working this way, and we don't have the results that we are looking for.'" 
    Ebrard initially looked within Mexico for a new strategy but was unable to find what he was looking for. "The people who wrote or proposed things about security, they are lawyers," he said. "The police never wrote books about policing, management of the police, or problems of the police. There was not even one text about the experience of the police in Mexico City." 
    Ebrard and his staff looked abroad, hoping to learn from the experiences of cities facing similar policing and security challenges-Palermo, Italy; Bogotá; and New York. Those consultations and the secretariat's internal evaluation produced a strategy that focused on information gathering, accountability, cooperative relationships between police and citizens, and more-efficient and smaller police units.
    López Obrador's fierce dedication to public security paved the way for many changes Ebrard had in mind. Immediately after he took office, López Obrador started daily, early-morning cabinet meetings on public security with 

Ebrard, the secretary of government, Mexico City's attorney general, the mayor's top legal adviser, and the director of prisons. The public security cabinet was the only cabinet with which López Obrador met daily.
    According to Ortega, who organized the cabinet meetings during 2003-04, the meetings frequently focused on the daily progress of public security and a rundown of the previous day's crimes. "We discussed the events of the night before," he said. "We sorted through them to find relevant events. If we identified something relevant, we separated it out in order to follow up on it. That's very important." He emphasized that the mayor always chaired the security cabinet meetings, which gave López Obrador an opportunity to personally monitor the daily progression of reform.
    "He was always here, and it was characteristic of him to give instructions directly," Ortega said of López Obrador's active role in the meetings. "He always assigned work."
    According to Ebrard, the meetings also represented an opportunity to enlist other government agencies in the effort to deal with Mexico City's security problem. "If you have a cabinet meeting, if you want to talk about safety issues, 22 agencies are thinking, 'Well, this meeting is not really about us,'" he said. "We tried to involve other agencies to think about safety. We said, 'You can do something to help safety, because safety is not a police issue; it's a government issue.'"
    That strong and consistent political commitment from the highest level of city government was crucial to the evaluation and monitoring of progress and to ensuring that the entire administration mobilized for the reform effort, according to Ortega. The meetings were intended to send a signal-from top police officials to cops patrolling the streets-that public security reform was López Obrador's top priority. Manuel Camacho Solís, a previous head 

of government of Mexico City, noted that the 
López Obrador administration hoped that members of the police force would begin to feel that their bosses were watching them.
	Ebrard realized early on that reorganizing the preventive police would require dealing with fundamentals like accountability. "We didn't have computers, so we could not verify every day who is where, who is responsible for what, what's your mission, which mission do you have every day, are you successful or not?" he said. 
    Ebrard stressed that more-complex and delicate issues-public trust and credibility-were equally important. "It's not possible to win a battle against organized crime in the city if you have your force isolated in front of society," Ebrard said, referring to Mexico City's history of distrust and animosity between citizens and the preventive police. "The first thing to do was to put the police force on the public scene, define their mission, and try to build a bridge between the community and the police. Otherwise, there is no possible solution. This was the starting point of the strategy by 2002."
	To begin chipping away at that distrust, López Obrador and Ebrard reached out to the business community, an unlikely ally for the left-wing PRD administration. Before Ebrard's appointment, López Obrador had begun building alliances with powerful members of Mexican society so as to secure and redevelop the city's downtown central historic district, creating the Consultative Council for the Rescue of the Historic Center of Mexico City.
	Ebrard expanded that effort to bring the business community into the administration's public security reform strategy by inviting prominent members of the business community to sit on a high-level citizen committee. The key, he said, was to recruit "people who really want to do something," including "people in charge of hotels in the city, the people who had suffered some kind of important crime against their family or their integrity." 
    Ebrard also stressed that the citizen committee had to be an integral part of the anticrime reform rather than "a decorative citizen council" that existed in name only. "You cannot invite the people to be a part of something if they cannot express freely the ideas they have. It's a risk, because they can say, 'Your strategy is not working.' But the worse risk is to lie to the city if the strategy is not working. You should change the strategy, not look for excuses."
    After recruiting the business community, Ebrard moved to build public confidence that the government was serious in its pledge to control crime, to fix the preventive police, and to underscore the administration's commitment to substantive, potentially radical change.
    "I started in 2002 to look for endorsement because the people hear each year, 'We are going to provide better security forces and strategies, and don't worry; we are working for you.' Really?" Ebrard said. "So there was a lack of trust between society and the authorities responsible for safety at the time. We invited foreign advisers . . . in order to say, 'This is a very serious strategy,' in order to build a new bridge of trust." 
    In 2002, a group of Mexico City businesspeople that included Mexican business magnate and billionaire Carlos Slim Helú agreed to pay former New York mayor Rudy Giuliani's consulting firm approximately US$4 million to design a plan for preventive policing reform in Mexico City. The group hoped that Giuliani, a conservative former prosecutor who had earned a reputation as a tough crime fighter while serving as mayor from 1994 through 2001, could help the security secretariat develop ways to quell Mexico City's crime wave.
    Giuliani's involvement cultivated support across the political spectrum. "Someone like Giuliani was quite accepted by entrepreneurial groups and classes, so for a government that looked to the left, having the advice of somebody accepted by the right, when all the attacks against the lack of security in the city were coming from the right, helped politically," said Camacho, who headed Mexico City's government from 1988 to 1993.
    The Giuliani consultancy's 2003 report recommended (1) focusing on improving "conflictive zones" such as Tepito-a low-income neighborhood a mile from the city's historic downtown and notorious for the trafficking of counterfeit goods, robbery, resistance to police control-and (2) taking a zero tolerance approach to informal businesses such as street vendors and windshield cleaners. The report recommended an information management system and a philosophy similar to the CompStat (computer statistics or comparative statistics) system that Giuliani had implemented in New York to map crimes, increase the accountability of individual policing units, and improve the allocation of resources. CompStat focused not just on the collection of statistics but also on the continuous improvement and development of the data. 
    With the release of the report, Ebrard began to set forth aggressive goals for the city's public security secretariat. "We think we can reduce the rates of serious crimes such as murder, rape, robbery, and kidnapping by 10% a year," he declared in August 2003.12
    As he sought to learn from outside examples, Ebrard also looked within Mexico for ideas. In mid 2002, he asked a retired army general, Enrique Pérez Casas, to join the team that would reorganize the police. Pérez Casas was not new to the challenges of police reform in Mexico City. He and Ebrard had first worked together in the 1990s under Camacho's mayoral administration. In 1990, Pérez Casas had tried and failed to implement a program inspired by Chile's Carabineros national police service. Pérez Casas said he believed that education and training-especially in values such as simple courtesy-was a fundamental reason for the success of the Carabineros. He reasoned that without such values, civil servants with close public contact, such as the police, could never be effective.
    Pérez Casas designed his 1990 plan as an incremental approach. He planned to create new police units whose members would enjoy improved working conditions, receive better and more training, and earn higher salaries. But he managed to get only two new police units up and running before the program collapsed in the face of powerful and unexpected opposition. "I made a mistake," he said. "I demonstrated that with 400 officers, it was more efficient than 80,000. And that raised the ire of the 80,000, and they started to tell me, 'This is your problem.' So I left." 
    Camacho's successor dismantled the program. "The old guard of the police allied themselves with the old guard of the PRI and again gave all the resources to the old guys in the police," Camacho said. "So this effort was completely lost." 
    As part of Ebrard's team, Pérez Casas had a second chance. Bolstered by stronger internal support, he proposed a solution that was nearly identical to his earlier approach. He would create Citizen Protection Units (Unidades de Protección Ciudadana, or UPCs), which would replace the preventive police in specific areas of the city. Ebrard and Pérez Casas planned that the UPC model would eventually operate in all areas of the city, thereby replacing the existing preventive police force.

GETTING DOWN TO WORK
    Just before Ebrard became head of the public security secretariat in 2002, López Obrador had negotiated between the SSPDF and the attorney general's office an agreement whereby the preventive police and the judicial police would divide Mexico City's 16 boroughs into 70 sectors-akin to police precincts. In each sector, Ebrard started to replicate López Obrador's cabinet meetings. He brought together coordination committees consisting of representatives of the police, local prosecutors, and the mayor's office that held daily meetings to share information and coordinate activities. The system enabled public security officials to deploy officers more precisely, assign responsibilities for supervision, and track changes in the nature and prevalence of crime. The coordinated area system also assigned to units certain outreach responsibilities for specific communities and mechanisms whereby members of the public could register complaints.
	López Obrador and Ebrard also copied at the sector level the citizen-council model they had used to enlist high-level outside support. Interested citizens could meet weekly with the police's sector chief, a representative of the attorney general's office, the local borough government, and the mayor's office. "In these meetings, citizens could present their grievances," said Ortega, who ran the security cabinet. "People could complain if, for example, they had caught a police officer in an act of corruption, or if services weren't on par." 
    Each day, a police representative from one of the 70 councils would attend López Obrador's daily security cabinet meeting to update the mayor's office on the situation in the particular sector. "The mayor's office heard every complaint," said Ortega, who helped each sector prepare its presentation. "Without a doubt, this was the period in which corruption decreased the most, because corruption is generated by a lack of supervision at the intermediary levels-independent of whether there are cultural issues. This demotivated people who could have been subject to corruption because they knew the people would be appearing in front of the mayor every so often and would be able to present evidence of corruption."


Creating Citizen Protection Units
    In November 2002, Pérez Casas began implementing the Citizen Protection Unit program under the subsecretariat for institutional development. He first recruited a staff from both the public and private sectors to help manage the program. Lorenzo Fernández Nieto, a retired army officer, was one of several ex-military colleagues he hired. A former member of the presidential guard, Fernández had been in the private sector for three years when he received Pérez Casas's recruitment call. By 2003, Fernández was one of six regional chiefs, overseeing 16 preventive policing units with the task of both reducing crime in his coordinated areas and converting the police to UPCs. 
    The UPCs were designed to operate in specific sectors of the city, developing relationships with the local communities. In exchange for better pay, training, and working conditions, UPC police worked under a more rigorous accountability system, which held them to higher standards of behavior. To measure performance, administrators created a point system that determined bonuses and promotions based on statistics the unit commanders collected each day. The indicators ranged from simple tardiness to changes in crime statistics in specific patrol areas and how much training the police officers completed. Those who achieved point goals earned financial rewards and promotions. Underperformance could result in the loss of their jobs.
    "Now there are clearer, institutionalized career advancement pathways accompanied by financial incentives," said Norberto Nava, who oversaw community policing. "New police officers know that a good performance record can get them promoted; this encourages them to take their jobs more seriously and to better protect the community." Fernández agreed: 


"When you have a future and you're earning money, you behave properly."
    The units were meant not only to promote effective law enforcement in crime-prone areas but also to address weaknesses in preventive policing in terms of workloads, pay, and training. At the time, Mexico City's police worked full-day shifts, with two days off between shifts. "The police in Mexico work 24 hours continuously," Pérez Casas said, referring to his observations of the system at the time. "No one can work like that."
    The work schedule for UPC police called for rotating, eight-hour shifts, with a day off every four days. Each unit comprised 400 police officers, with 100 on duty at any given time. 
    Higher salaries also served as a cornerstone of the UPC program. "The police come from social classes that are very poor," Pérez Casas said in describing the problem. "They're police because they want to eat, not because they have any desire to reinvigorate society. They're not treated well, and they don't have money for food-basic things everyone needs-and it obliges a person to become corrupt."
    UPC recruits who passed entrance exams and completed their training would receive a monthly salary of 8,500 pesos (US$787 in 2003)-more than double the roughly 4,000 pesos (US$370) paid monthly to the preventative police. 
    Changes in recruiting and training were fundamental elements of the plan. To develop a new curriculum for the UPCs, Pérez Casas and his team began working with the Technical Institute of Police Training, which handled instruction for all incoming police recruits. They tried to attract younger recruits, who the team believed would be more idealistic and have less exposure to the kinds of corruption that infected Mexico's police. They raised the minimum education requirement to completion of secondary school. They developed a mandatory exam for new recruits. And they updated the training program for those who passed.
 	"Training was no longer that simple," said José Alfredo Carrillo García, who was director of preventive-policing actions at the public security secretariat. "Before, they had been told only how to hold and shoot a rifle and to march and they talked to recruits about what the police force was, and that was it. . . . We tried to teach new things like human, constitutional, penal rights as well as criminology." In 2003, Ebrard persuaded the Legislative Assembly to fund the new training system by pointing out that 90% of the police had not received official training from either the Technical Institute or Mexico City's National Autonomous University.
    Attracting qualified recruits initially proved challenging, as the UPC program had the same entrance requirements as the country's Federal Police but offered a lower salary. Eventually, Pérez Casas and his team lowered entrance requirements from requiring secondary school completion to a commitment to complete secondary education while in the UPC system. (In Mexico, only completion of the first half of secondary school, referred to as secundaria, is mandatory; the second half, or preparatoria, is optional.)
    Pérez Casas said he felt strongly that the training process should not mix new recruits with police veterans who might be tainted by corruption. "The police officers who had experience-those were my problem," he said. He decided to graduate entire units of 400 recruits together from the Technical Institute.
    Pérez Casas and his team added other bonuses that enlisted families in encouraging strong job performance by their main breadwinners. High achievers in the UPC became eligible for housing assistance and educational aid for their children. "I didn't have to take care of them," Pérez Casas said. "Their families took care of them because they knew: 

'Dad, I don't have a scholarship, and it's because you're not doing well.' The wife would say, 'We're going to lose our house.'"
    As time went on, Pérez Casas and his team recognized that they needed not only to encourage good behavior but also to reduce corruptive influences. For example, a police officer could make nearly a month's salary by seizing guns and selling them on the black market. So, to police who seized illegal guns, the UPCs began to give bonuses that were higher than the guns' black-market values. Fernández said official gun seizures jumped from around 300 firearms a year to 3,800. "Before, police officers would keep the gun and sell the gun on the black market, and the thief would not say anything because it was also in his best interest," because the thief would avoid a weapons charge, he said. "Now thieves caught with guns are getting five or six years in jail." 
    Rather than try to retrain and reorganize the existing preventive police force, Ebrard and Pérez Casas adopted a system of gradual replacement. "The goal was to have the entire police force be assimilated by the UPCs in about 10 or 12 years," Carrillo said. They focused first on implementing the units in areas of higher crime rates. Because the renewal of Mexico City's historic downtown was a political priority for the López Obrador administration, Pérez Casas and his team began implementing the first UPCs in that area. 
    At first, there were concerns that the local borough governments would resist the new model, but Pérez Casas found the opposite to be true. "People [in the borough governments] were complaining: 'I want one here, I want one here,'" he said. "Ultimately, I had to comply with what the leaders of the boroughs wanted." Although Pérez Casas said demand for UPC units was high, the replacement of existing preventive police was gradual. By the end of 2004, there were only six fully functioning UPCs.
Working with the rest of the preventive police
    Not surprisingly, many of the old-guard police resented the higher pay, better benefits, and improved working conditions enjoyed by incoming UPC recruits. "There were frictions between the old police and the new ones because of the higher salaries," Fernández said. 
    To speed up replacement of existing preventive police units with UPCs and to eliminate underperforming or corrupt members of the security service, Ebrard adopted a two-pronged strategy. "The first was to offer to the oldest officers in the force a good idea of retirement, because in Mexico traditionally you didn't have a decent retirement plan, so you tried to stay on the force as long as possible to avoid lack of money or support," Ebrard said. He instituted a system of voluntary retirement, particularly for higher-level and older members of the police. Rather than dismiss officers and risk a repeat of the 1998 police strike, the public security secretariat retired them with honors and increased pensions. "This eliminated 70% of the people who could not adapt to the new structures," Fernández said. 
    Those who refused voluntary retirement and could not meet higher efficiency standards were discharged. From March 2003 to February 2004, for example, the SSPDF fired 1,270 officers and internally suspended an additional 1,154.13 Those who could meet efficiency and new training standards were given opportunities to enter administrative positions or, on rare occasions, one of the UPCs.
    "When we started, it was risky but finally worked, because you can win a battle by time," Ebrard said. "The message is, if you do the job, you have the right to be on the force and you have the support of the force."
    For police who remained on the force, Ebrard and his team designed a system to encourage higher performance and to weed out those who lacked the desire to raise their personal and professional standards. "The other [strategy] was to increase the standard average of requirements for the police and their performance," Ebrard said. "We started to measure each officer and, naturally, a lot of them preferred not to be on the force with those kinds of efficiency requirements." In an interview with a scholar, a member of Ebrard's team said Ebrard "worked on the assumption that 20% of the police were highly corrupt, 20% were very honest, and the remaining 60% could be improved or better managed with organizational reforms."14
    Among other measurements, officers were tracked for absenteeism, completion of new training, and arrests that resulted in convictions. In an effort to promote more-active policing, those who made successful arrests of thieves were eligible for financial bonuses. "If the person who was arrested went to jail, we paid the officer," Carrillo said. The bonus for arresting someone who stole a car, for example, was 3,000 to 5,000 pesos (US$265 to US$442 in 2004)-the equivalent of about a month's salary.
    Although the financial incentives were effective in boosting the number of arrests, critics pointed out that the plan also led to some abuses. "The police were offered benefits for detentions, and they made illegal arrests," said Mario Patrón, chief investigator of the Mexico City police for the Federal District Human Rights Commission.
    Carrillo said police leaders tried to combat the problem: "If in the process the person proved that he was falsely arrested, we sanctioned the police officer to prevent the idea that the police officers could take anyone and put them in jail." In these cases, Carrillo said, a committee of higher-ranking officers, members of internal affairs, and lawyers from the public security secretariat would review the arrests and potential punishments. No data were available regarding the committee's level of success in deterring officers from making wrongful arrests. The public security secretariat eventually abandoned the bonus program on the recommendation of the Federal District Human Rights Commission, Patrón said.

Building better police-community relations
    In the sectors of the city where they worked, UPC police patrolled with the explicit mission of interacting with residents of their assigned neighborhoods both to improve relations with the general public and to establish personal relationships that might become lines of communication and intelligence. "We sent them out to knock on every door to say, 'I'm Lorenzo 
Fernández Nieto. Here's my contact information. I'm patrolling your area, and I'm here to protect you,'" Fernández said. 
    However, many of the strategies met with initial resistance. "Sometimes it was very difficult because we got to the door and knocked, and many of the citizens wouldn't answer," Carrillo recalled. "Sometimes they would call the police and say, 'Hey, there's a police officer knocking on my door. What's going on?'" 
    Pérez Casas said at least part of the problem stemmed from the fact that the UPC police refused to go along with the law bending that had become common in the past. "When I entered [a neighborhood] with the UPC, the citizens gave me a hard time," he said. "They would say, 'The other ones were better. With 10 pesos, they'd help me move my car. And with your guys, all they do is give me tickets. We want the other ones back.'"
    In 2004, Ebrard deepened his emphasis on community relations by creating the Neighborhood Police (Policia de Barrio). Under the original design, the UPC police were intended to stress citizen relations, but the short-term emphasis on lowering their sector's crime statistics left them with little time to build rapport with residents. "The UPCs and the Neighborhood Police started as one unit," Carrillo said. "But we didn't have the reaction we expected from the citizens. It was difficult to get citizens to open their doors to the police. We realized this wasn't working, so we had to make a lot of changes."
    With the goals of improving citizens' access to police and bolstering community credibility, the Neighborhood Police drew from the non-UPC members of the preventive police force. The changes were sometimes subtle. Although patrol beats and training remained largely unchanged, their evaluation criteria became stricter. For example, officers had to obtain citizens' signatures to verify they had completed their assigned patrols. Neighborhood Police officers participated 
in the meetings of local citizen committees, which served as intermediaries between the public and the security secretariat. 
    "The objective was to bring the police force closer to the average person," Carrillo said. "The UPC would patrol all around the sector, and the Neighborhood Police would go to the different houses to knock on their doors and ask people about the problems, about the environment, about the information. We tried to increase contact between the police and the citizens. The principal difference was that the UPCs were only patrolling, and the Neighborhood Police were directed to face the citizens."
    By 2006, the Neighborhood Police numbered 1,836 members, all of them drawn from the preventive police but without specialized training. "It just meant that the police assigned to a specific area had the responsibility to get to know the local community leaders," Carrillo said. "The Neighborhood Police was only an idea. It was only meant to make the police force more approachable."

Using technology to fight crime
    When Ebrard joined the public security secretariat, the organization had been relying heavily on a decentralized, paper-based information system that made it difficult-if not impossible-to compare and analyze crime information from across the city. While working on the longer-term problem of reorganizing the preventive police and integrating the use of technology into public security, Ebrard implemented several ad hoc measures aimed at improving police responsiveness in particular situations. In August 2003, for example, in response to the problem of bus muggings, Ebrard began installing panic buttons throughout the city bus system with the promise of a four-minute police response time.15
    In November 2003, the security secretariat began developing the Police Information System to aggregate, compile, and analyze crime information from across the city. A private contractor developed a system along the lines of New York's CompStat program. The secretariat rolled out the system to each of the 70 coordinated areas, prioritizing those with higher crime rates. Each coordinated area would submit its statistics and reports to secretariat headquarters daily, enabling Ebrard and his team to track progress in real time. 
    "I estimate that every day, we had 500 to 600 registries being uploaded about the people who were detained, the calls made to report crimes, among other things," Carrillo said. At the same time, the secretariat created programs to track media reports of crimes and officer conduct.
    The data enabled the secretariat to deploy resources more efficiently and to change patrolling patterns to fit areas and times of high crime. The Police Information System "made it possible for us to know which were the areas with the highest crime rates and gave us the capacity to act and prevent those crimes," 
Carrillo said. But early efforts to use the new system encountered predictable snags: input errors diminished the initial effectiveness of the program because many police had only limited experience with information technology. "It was difficult because not many people were comfortable using computers," Carrillo said. "I think that initially, we lost a lot of information because not everyone could enter the information to the database. It was a hard learning curve, and a lot of mistakes were made." 
	As those initial problems faded, the data-driven approach enabled Ebrard to hold the chiefs of individual sectors to account for their units' performance. "You knew from the statistics what was out of place, so you could know or check what was happening there," said Leticia Bonifaz, who joined the SSPDF in 2004 as coordinator of special projects. "The sector chiefs in sectors that improved received prizes, and the sector chiefs in areas that got worse were called before the secretary of public safety each Monday to explain. And when it was clear that someone was no longer effective [on the job], they dismissed that officer, unless the officer who was the cause of the irregularities was found. There was shared responsibility with the sector chief, who had to be able to control the officers. You could make changes or adjustments, but the statistics from the report told you that something was happening."

Focusing on the Central Historic District and Tepito
	Ebrard was inspired by the experiences of Leoluca Orlando in Palermo, Italy, with whom he consulted early in his term as secretary of public security. Ebrard knew that securing Palermo's historic city center had been essential to Orlando's success in reenergizing downtown economic growth and fostering a broader sense of security in the city as a whole. Ebrard and his team knew they would need to focus on high-visibility areas of Mexico City to demonstrate the viability of the policing reforms and to build public confidence in the rule of law.
    In 2000, the downtown historic district had one of the highest crime rates in Mexico City: 24.5% of total crimes that year.16 It was also a center for informal and often illegal businesses such as prostitution and the sale of counterfeit and other illicit goods. Focusing on visible steps to clean up the historic district, Ebrard and his team took aim at the informal street vendors who clogged the streets. 
    "In the central historical district, you couldn't even walk because there were 30,000 people selling things in the streets," Fernández said. In 2000, there were an estimated 120 street vending organizations in the downtown area, comprising 30,000 members. 
    Another problem area was nearby Tepito. The low-income neighborhood was a staging ground for organized criminal activity throughout the city, particularly drug trafficking.17
	Reasserting a security presence downtown also meant installing surveillance cameras throughout the area, a move that Ebrard said had been inspired by such a system in London. The López Obrador administration credited the installation of security cameras with a 30.8% decrease in crime in the area from 2003 to 2004.18
    Ebrard drew inspiration from former New York City Police Department commissioner William Bratton, who subscribed to the broken-windows theory of policing. That strategy held that restoring order to crime-ridden areas required aggressive policing of minor offenses such as illegal street vending, disorderly conduct, public drunkenness, and loitering. The thinking ran parallel to the zero-tolerance response former New York mayor Giuliani's team had advocated in its report.
    To establish the authority to use that kind of approach, Ebrard and Bonifaz pushed for the Law of Civic Culture, which the Federal District legislature passed in mid 2004. Inspired by Giuliani's earlier zero-tolerance recommendations, the law criminalized many types of informal street businesses and increased the penalties for public intoxication and drug possession. "Among Giuliani's recommendations were to reinforce preventive measures," Bonifaz said. "This meant paying a lot of attention to minor civil offenses-for example, graffiti."
    The strategy prompted pushback, particularly from judges whose dockets might become clogged with petty cases. "Certainly, it was difficult to convince the legislature, but the judiciary was more resistant because [the change meant] more work for them," Bonifaz said. To assuage judges' workload concerns, a companion proposal removed minor cases like vehicle accidents from their dockets, she said.
	
Handling an abrupt transition
    Ebrard and his team were two years into their implementation of police reform when disaster struck. In November 2004, a mob in the Mexico City borough of Tláhuac mistook three undercover federal agents for child kidnappers. Two of the three were beaten and burned alive in front of television reporters. Media outrage following the incident focused on the failure of the Mexico City's preventive police to stop the violence or mount a rescue attempt during the two hours the federal officers were held. In response, Mexican president Vicente Fox fired both Ebrard and the commissioner of the Federal Preventive Police.19 López Obrador initially defended Ebrard but later accepted Fox's decision.
    López Obrador appointed his adviser, Ortega, to succeed Ebrard as head of the security secretariat. Ortega, an engineer and former minister of transport for the city in the late 1990s, had a technical rather than a security background. However, Ortega had provided important technical support for the security cabinet meetings and advised López Obrador on security issues. 
    For the most part, Ortega kept Ebrard's team and reform plan in place through increased focus on technological improvements-a policy that smoothed what could have been a difficult transition. "With [Ebrard], we started solving some things, and when Ortega came, we were able to use all the information we collected before to make the programs successful," Carrillo said. "Continuity was important."
    In addition to technological skills, Ortega brought with him a strong appreciation for the informal underpinnings of Mexico City's society. "Ortega had the advantage of not only being a very competent engineer . . . but also knowing the social network of the city," Camacho said. "He had the capacity to deal with those social groups in highly conflicted sectors and the capacity to look to new ideas and new technologies as a very competent engineer and organizer."

Broadening the strategy
    The loss of his post as secretary of public security was not the end of Ebrard's political career. In 2006, after López Obrador left his position to run (unsuccessfully) for president, Ebrard was elected mayor of Mexico City, helped by strong public support for his security reforms. He took office in November 2006 and immediately reappointed Ortega and Pérez Casas, a clear sign that the security strategy remained a priority. 
    Ebrard continued López Obrador's tradition of daily security meetings. "There really were no changes in the strategy that I took over from Ebrard, and later, he became the mayor, so there was a lot of continuity," Ortega said. 
    Even though he remained committed to the same basic security reform strategy, Ebrard began to focus on an approach that extended beyond the police. "We have a lot of things that it is impossible to do just in the police force," he said. Concerned about the roles of teenagers and young adults in violence and criminal gangs, for example, Ebrard began offering scholarships and other incentives for students who remained in school and improved their grades. He also began in earnest removing street vendors from the downtown historic center of the city, a process he had planned while secretary of public security. Although Fernández said he and his fellow officers were able to move street vendors out of the downtown areas "without throwing a punch, by talking," other reports indicated a prolonged effort. The crackdown, which began in 2007 and increased in intensity through 2008, detained thousands of vendors-often on charges of interfering with traffic, because the selling of many types of goods in the streets was not otherwise against the law.
    Because legitimate street vendors were not given any alternative places to sell their wares, many moved to less-safe areas of the city, and even those who sold legitimate goods began to operate furtively, fleeing when police arrived, Rodrigo Meneses Reyes, a scholar of law and sociology at Mexico's Center for Research and Teaching in Economics, said. However, he added, "for the majority of citizens, the removal of the street vendors was viewed positively, because the space looked cleaner and was easier to deal with."
	Inspired by his earlier research into the experiences of Bogotá and Palermo, Ebrard expanded the strategy used in the central historic district, and along with Ortega and Bonifaz, now his top legal adviser, began to "retake" parts of the city that were known as hotbeds of certain crimes. "[Ebrard] said we have to seize property in Tepito, the worst part of the city, in order to have a presence there," Bonifaz recalled. 
    In February 2007, police surrounded a city block notorious for drug trafficking and the sale of stolen property. "The area was seized and demolished, and now there are a school, a swimming pool, a child care facility, and a space for children, all in the center of Tepito," Bonifaz said. "Now this area, in which there used to be only crime, is an educational space that people can go to."
    Following up quickly on their success in Tepito, Ebrard, Ortega, and Bonifaz planned a second raid. On a quiet public holiday in March 2007, they loaded police into buses and took them to an area known as La Ford in the borough of Iztapalapa, full of small shops known to sell stolen car parts. The police seized two hectares (five acres) of property and closed down the illegal operations. "Everything was demolished, and in its place a health center, a child care center, and a disabled persons center were built, as well as a park in the back," Bonifaz said.
    This was an expensive strategy because laws required the city to compensate the documented owners of the land and other legally obtained property. In response, Bonifaz began working on the Ley de Extinción de Domino, or Law of Asset Seizures. Modeled on a statute enacted in Colombia, the proposal would allow the Federal District government to seize property involved in the perpetration of a crime-without having to compensate the owners. 
    The proposed law required changing Mexico's constitution, which guaranteed the right to compensation in cases of government seizure. Bonifaz lobbied Mexico's federal government with the argument that the law would be useful nationwide. "The fact that this law could also be used in other states was what convinced them," she said. "We thought about the law in terms of local crimes, but many federal crimes-especially organized crime and narcotrafficking-could also apply." 
    Following passage of the constitutional amendment, Mexico City's government enacted its version of the law in June 2008, and the federal government followed suit the next year. "During [Ebrard's] administration, there were more than 130 cases of forfeiture of assets after 
the law was enacted, the majority of which were due to kidnapping, vehicle robbery, and human trafficking," Bonifaz said. "If the property is sold, a percentage [of the money from the sale] goes to support the Secretariat of Public Safety, the attorney general, and the victim."

Continued progress
    Under Ortega, the public security secretariat moved forward in a number of areas. Ortega set a high priority on digital monitoring and accountability to improve police effectiveness. Working to speed response times, Ortega established a 24-hour centralized emergency police telephone line and equipped patrol cars with GPSs that enabled the secretariat headquarters to track locations.
    By 2008, the public security secretariat had completed installation of the Evaluation and Performance Center, which had adopted a version of New York's CompStat system. Policy staff, police commanders, and midlevel officers in each of the sectors used the system in their daily meetings to review and analyze crime data. "Every day, one of the 70 'coordinated areas' was evaluated with the chiefs of every sector in that coordination, as well as with the prosecutor in charge of that area," Ortega said.
    Inspired by the success of cameras in the central historic district, Ortega's administration accelerated the installation of cameras in high-risk areas. In 2004, 150 cameras were monitoring roadways. In 2005, the public security secretariat spent 6 million pesos (US$550,200 at the time) to install a network of cameras throughout all 16 boroughs. And by 2007, more than 4,000 cameras were operation. 
    Ortega also oversaw a rapid expansion of the UPC program. Under Ebrard, Pérez Casas's 

team had established five UPCs. Under Ortega, they created 27 more, hiring thousands of new police officers. "He's a real mover, a creator," Fernández said of Ortega. "He really believed in the project."
    Ortega also set a priority on phasing out senior commanders in order to bring in younger officers who would help change the institutional culture of the police. He ramped up the voluntary retirement program and increased dismissals of underperforming commanders. "When I entered [as secretary in 2004], the average age of commanders was 72 years," Ortega said. "When I left office, it was 35." As it had under Ebrard, the public security secretariat held events to retire officers and give them awards, medals, and severance checks. 
    Ebrard and Ortega took aim at crimes that affected citizens' daily lives. One team focused on car thefts and worked with insurance companies to identify stolen vehicles and gain access to the companies' registries of stolen cars. The two also assembled a team focused specifically on thieves who targeted buses and truck drivers in areas of high crime. A third initiative took aim at bank robberies.
    Ortega continued to tackle sources of corruption. Tow truck operators had become major sources of extortion, Ortega said, because they would confiscate cars and then demand bribes for their return, circumventing the official system of impoundment fees. Ortega installed cameras on official tow trucks and required payment of fees by credit card. "We turned complicated things into very simple processes," he said. 
    Ortega also focused on internal corruption. "The police here had a program where they were selling their bullets to people who were selling the bullets to criminals," Ortega said. His administration began tracking the registration numbers of police weapons and stamped each bullet with identifying markings.

OVERCOMING OBSTACLES
    López Obrador, Ebrard, and Ortega's 

implementation of security reforms in Mexico City demonstrated their ability to deal with obstacles that had foiled earlier reform attempts: To handle a resistant old guard within the police, Ebrard and his team worked to make retirement a more attractive option. To weather an abrupt transition when Fox dismissed Ebrard, López Obrador appointed Ortega, the head of his security cabinet, to ensure strategic continuity. And when Ebrard became head of Mexico City's government, he kept Ortega on and retained the security secretariat.
    During Ebrard's mayoral administration, however, systematic problems within the preventive police continued to pose challenges. Most prominently, many arrests failed to lead to formal investigations by the judicial police or to prosecutions. "The impunity was a problem because every year, the police in the Federal District arrest 9,000 people and only 10% get sentenced to jail," Carrillo said.
	In 2006, Bonifaz began studying failed cases to see whether she could spot a common thread. She soon realized that judges were dismissing cases based on problems with the application for criminal complaint, which was a form completed by arresting officers in order to begin an investigation by the judicial police.
     Bonifaz said she concentrated on cases involving drug dealing. "I reviewed 100 cases to find the mistakes, and the surprise was that all of them were the same," Bonifaz said. "I told Joel Ortega, who was the chief of police, that he doesn't have a single drug dealer who is left-handed because all the reports were the same. All of them started with, 'While doing the patrol rounds, we saw some people who looked suspicious. Upon approaching them, we discovered that in the right pocket of their pants they were carrying'-here was the change-'a green herb that looked like marijuana,' or it could be 'something white that appeared to be cocaine.' All of them were the same. No one carried anything in a shoe or a hat-always 'in the right pocket of their pants a green herb that looked like marijuana or a white powder.'" 
    Bonifaz also found that judges were throwing out cases because suspects had been mistreated and confessions coerced. "The issue involved preventing torture, which was the most delicate problem," she said. "It was a very common practice for the police to have already extracted a confession before arriving at the Ministry of Public Safety." Sometimes, she added, police officers had videotaped the confessions, and the recordings showed clear evidence of abuse.
    Bonifaz said the problems revealed police ignorance of required procedures and lack of appropriate training. The public security secretariat began working to train officers better in arrest and evidence procedures and, with the Federal District Human Rights Commission, to explain the rights of suspects. "The training sessions that the human rights commission held for the police were trying to teach how to do the job, how to make arrests, how to send cases to the authorities, to the judge," Carrillo said. 
    The human rights collaboration made slow progress because many police veterans viewed procedures designed to protect human rights as favoring criminals. "They see human rights as the enemy," Carrillo said. "It got a little better; there was something good about the training sessions," he added. "Still, now, it is hard to make the police officers pay attention to the human rights programs. But they know they have to go; they have to take the course." He expressed confidence that generational change would make the ultimate difference. "The new generation that are getting into the police, they have a new idea" about human rights, he said. 
    Despite those efforts, alleged human rights violations remained a serious problem for the Federal District police. Patrón, chief investigator of the police for the Federal District Human Rights Commission, attributed the ongoing problems to weak internal and external controls. "One of the things we've identified that's missing is a body within the police to conduct investigations, because right now they have internal affairs, but it doesn't have investigative capacity," he said. "They need an independent organ to investigate the police."
     Meneses echoed that assessment: "If you do not have external checks and balances, it is impossible to improve a situation that is produced by the same police."
    
ASSESSING RESULTS 
    Decreased crime in Mexico City could not be attributed solely to the preventive police reforms, but under López Obrador and Ebrard, the city showed clear improvement. When López Obrador came into office in 2000, crime rates in Mexico City had risen dramatically during the 1990s, and the government was struggling to control the police. 
    Following a significant drop from their peak in 1999, overall crime rates declined steadily during the years of the police reforms until 2006, when they began increasing. While the overall crime rate did not show sustained improvement, the López Obrador and Ebrard administrations did achieve success in specific target areas: Robberies, particularly violent robberies, declined during both administrations. Vehicle thefts dropped from an average of 119.25 per day in 2000 (already down from a high of 160.22 in 1997) to 72.82 per day in 2006.20 According to the administration, López Obrador and Ebrard had achieved a 30.8% reduction in crime in the central historic district.21 And police efficiency levels had risen by 10% from Ebrard's initial measurement in 2002, according to public secretariat records. Much of that data, however, was not readily available to the public.
    It is worth noting, when analyzing these statistics, that both Ebrard and Ortega attempted to improve the percentage of citizens reporting crimes.
    In 1999, in a survey by Transparencia Mexicana, 90% of Mexico City residents reported having little or no confidence in the Federal District police, indicating that no more than 10% had any significant confidence in the police.22 A separate study from 2006-11 found that 28% and 42% of the population reported confidence in the police.23
    Still, with more than half the populace skeptical of the police, reformers clearly had more work to do. "The perception of insecurity continues to be very high in Mexico City, and that's something that systematically runs against all of the leaders' speeches," Colegio de México professor and policing expert Arturo Alvarado said. "That's the stain on all these programs." Mexico City, Alvarado added, has managed to avoid much of the drug cartel and other criminal gang warfare that plagued cities like Monterrey and Juárez. "In the early '80s in Mexico City, the most important mafia was the Mexico City police," he said. "So, that shift meant a lot of changes inside the police organization."
    Some observers, however, argued that corruption remained a significant hurdle for the police. "I think that Joel Ortega and Marcelo Ebrard had good intentions with that, and it would be a nice and effective policy in another part of the world, but here in Mexico City, the problem is that the high levels of the police are those who are in charge to sanction the misconduct of the same police officers," Rodrigo Meneses Reyes, a scholar of law and sociology at Mexico's Center for Research and Teaching in Economics, said. "The mechanisms of institutional control never got better."
    Before their abandonment under Manuel Mondragón y Kalb, who succeeded Ortega in 2008 as head of the public security secretariat, the UPCs existed in 32 of the 70 coordinated areas. According to one study in January 2008, 37% of interviewees said the UPC officers were less corrupt than the preventive police.24 Fifty percent said they were just as corrupt. Federal District Annual Crime Rate, 1999-2011 (see figure on PDF). Although the public had a better image of UPC officers, 73% of interviewees did not know the difference between UPC and non-UPC police.25 The same reports raised serious questions about the effectiveness of the Neighborhood Police program, finding that 97% of survey respondents did not know the police patrolling their neighborhood.
    In 2008, Fundar, a Mexican nongovernmental organization, and Juan Salgado, scholar of police reform at Mexico's Center for Research and Teaching in Economics, released an independent evaluation that praised the UPC program. At that time, the UPCs operated in 22 sectors in Mexico City. "The UPCs were the most far-reaching police reform project in Mexico," Salgado said in an interview. "The SSPDF turned 36 [sectors] into UPCs despite formal and informal resistance within the police. That success had to do with the continuity of General Enrique Pérez Casas. Also, that reform matched civil society groups' and human rights groups' demands for clearer structure in the police, checks and balances, and better salaries. The UPCs achieved many of their goals-at least structurally."
    "The report establishes and maintains Citizen Protection Units in SSPDF are one of the few examples that meet the agendas of civil society organizations, the [Federal District] government, and police matters," Fundar concluded. "In the capital and other parts of the country, the UPCs represent one of the most outstanding models of modernization of local police operation, especially in terms of police approach to citizens."26
    Fundar also found that the UPC officers appeared to be more dedicated to their work, which had been one of Pérez Casas's priorities. "Fundar's team identified a team spirit and a special motivation on the part of most of the officers of the UPC that is hard to find in other police agencies," the report stated. "Although receiving higher wages should have an influence on the motivation of these officers, we should also recognize that there are elements of leadership and shared mission that create cohesion and make them feel proud of their work."27
    The average response time of a UPC was two to eight minutes-far faster than that of the rest of the preventive police. The UPCs also were more effective at prioritizing higher-impact crimes. When the program began in 2002, 39.2% of the incidents that the preventive police referred for investigation were automobile accidents. By 2007, that figure was down to 18.2%, as the focus shifted toward robbery and other crimes. Fundar also found that in the years the UPCs were in operation, thefts decreased from 95,617 in 2003 to 78,264 in 2006.
    From 2002 through 2007, the UPC program cost the public security secretariat a total of 2.064 billion pesos (US$188.7 million). Each UPC installation cost the Federal District government 76 million pesos (US$6.9 million) and the local borough government 18 million pesos (US$1.6 million).28 During that period, Fundar found that the UPC budget caused no change in the secretariat's roughly 9-billion-pesos (US$822.6-million) annual budget, which remained stable.
    That said, Fundar identified serious flaws in the program. The report said many UPCs were underfunded and lacked promised equipment. The researchers noted that no UPC had a formal system to receive citizen complaints and that UPC officers frequently did not participate in the weekly citizen meetings (they were not required to do so). Furthermore, Fundar found human rights violations remained a serious problem. 
    The effort to clean up the central historic district and Tepito was a high-profile success for López Obrador and Ebrard. Although reestablishment of security in the downtown led to a revitalization of the area, the policing efforts may have simply driven illicit activity elsewhere in the city-or just over the Federal District's borders into the greater metropolitan area. Criminal organizations "preferred to go to the neighborhoods in the state of Mexico, rather than here," Ebrard said. 
    Many sectors were able to improve their performance, as measured by the crime statistics system and other secretariat indicators, but outside observers raised questions about incentives the system created, particularly the targets set for arrests of thieves. "The SSPDF's adoption of the recommendations made in the Giuliani report have led to domestic policies of police performance appraisal based on meeting quantitative rather than qualitative criteria. This represents a perverse incentive for agents to fulfill their work at the expense of respect for the rights of citizenship," Fundar concluded in 2008.
    In 2011, the public security secretariat ended its use of financial incentives for arrests, a move that, according to Patrón, was based largely on a recommendation by the Federal District Human Rights Commission. However, he added, "In practice, these types of things still exist. It's become a perversion of justice."
    Mexico City received international attention for its handling of crime. In 2012, the US State Department rescinded the critical-criminal-level warning about Mexico City that it had issued in 1998, thereby signaling significant progress in the city's law enforcement situation. Mexico City also received attention from the international press, which dubbed the city's handling of crime a "miracle."29

REFLECTIONS 
    From 2001 to 2008, Mexico City's public security secretariat, led by Marcelo Ebrard Casaubón and Joel Ortega, focused on implementing a new model for preventive policing in the city by using technology to increase police force efficiency and combat the city's crime wave. Ebrard and his team built their strategy with an eye toward Mexico City's earlier failed attempts at police reform and inspired by the successes of reformers in New York, Bogotá, and Palermo, Italy. 
    Whereas the long-term impact of the preventive police reform under the administrations of Andrés Manuel López Obrador and Ebrard remains to be seen, their creation of specialized Citizen Protection Units and Neighborhood Police, their implementation of technological advancements, and their coordinated efforts to ensure lawfulness in the city's historic center and other high-profile areas of the city represented a sustained effort to combat rising crime rates and strengthen the rule of law. The reform benefited from high-level and consistent political support, a shared commitment to the importance of the reform, and continuity between administrations.
    Accountability was a hallmark of the reform. The daily security cabinet meetings that López Obrador started were instrumental in tracking progress and holding reform participants to account for what they did and didn't do. "Your number one priority was to be there [at the meeting]," former Legal Advisor Leticia Bonifaz said. "There was the commitment to making sure that everything that was planned actually happened, because we Mexicans are good at planning but not very good at following through. How do you know whether people are actually doing things or not? Because every day, they're asking you, 'What have you done?' or 'How is it going?' or 'Are you in the same place?'"
	Cultivating strong support was another linchpin of the effort. Ebrard pointed to the importance of data and the citizen councils in holding police management accountable. "It's a matter of choosing the correct people but always having a very important supervision system, an independent system. And if you can invite citizens, all the better."
	The continuity of policy between López Obrador and Ebrard as heads of government 

and between Ebrard and Ortega as secretaries of public security played a key role in effective implementation of the police reforms. Unusual in Mexico, that degree of sustained focus on one set of reforms led to greater institutionalization of the changes and steeled the public security secretariat to weather an abrupt leadership change in 2004.
	Ebrard and Ortega, however, were less successful in institutionalizing that commitment throughout the public security secretariat. Under Ebrard and Ortega, "levels of corruption went down, but this was the result of the attitude of the secretary of police and his high-ranking officers; it's not institutionalized," Mario Patrón, chief investigator of the Mexico City police for the Federal District Human Rights Commission, said. "They haven't revised the institutional engineering of the police."
	Patrón and other observers-including Meneses-also raised serious concerns about the lack of strong internal and external controls on police misconduct. "I think there are a lot of good intentions in the police officers of the city, but they are probably also scared of what they'll find in the institution," Meneses said, arguing that police leadership had not gone after internal corruption strongly enough.
	Mexico City's drop in crime rates had causes other than the police reforms under López Obrador and Ebrard, but by the end of Ebrard's term in 2012, Mexico City had transformed from one of the least-secure cities in the country to one of the country's safer metropolises. "If you compare what is happening in Mexico in the past 8 or 10 years both in Mexico City and at the federal level, Mexico City seems like a paradise," Alvarado said. He questioned, however, whether it was due to real improvement within the city or increasing instability elsewhere. "If you analyze the trends in crime in the capital, this paradise fades," he said.

EPILOGUE
	In mid 2008, the public security secretariat was again rocked by a tragedy that destabilized its leadership. This time, it weathered the transition less strongly. In June, Mexico City police raided a nightclub on suspicion of illegal alcohol and drug sales. In a stampede caused by the raid, 12 clubgoers died. Several of the people detained in the raid filed complaints of police brutality and sexual assault. Of the 120 police who conducted the raid, 30 were UPC police who had received community policing and human rights training. All had clean records prior to the raid.
    The aftermath of the raid raised questions in the media as to the effectiveness of the UPC program. The ensuing backlash also led then president Felipe Calderón to remove Ortega as head of the public security secretariat. Ebrard chose Manuel Mondragón y Kalb as Ortega's replacement. Mondragón had worked in both the secretariat and the attorney general's office before becoming the new secretary-experience that appealed to Ebrard. "We needed someone like this man because he can organize three very difficult things to have in the same person," Ebrard said. "First of all, management capability; second, capacity for communication with the people, . . . and third, military formation in order to have the force organized."
    With Mondragón, Ebrard continued much of the work he had begun with Ortega. 
    In 2008, Ebrard and Mondragón created Centers of Attention, which were specialized agencies aimed at improving police engagement with citizens in general and marginalized groups in particular. The centers also had the task of tracking reports in their areas and allegations of police abuse, which they forwarded to internal affairs. 
    In March 2009, Ebrard expanded the security camera program throughout the city by ordering a $460-million Safe City surveillance system. "It's costly, but it makes sense," Ebrard said of the system. "If you compare 2006-12, you have a reduction in the city by about 40% in the most important crime index for the people. Part of this was because we had the video cameras."
As head of the public security secretariat, however, Mondragón distanced himself from the UPC model in the wake of the nightclub raid scandal. He brought in a new team, which replaced many of retired army general Enrique Pérez Casas's staff. Pérez Casas left in 2009, saying he had been pushed out. "The work of 38,000 well-trained people was erased by Mondragón," he said. 
    In 2009, Mondragón began working on a new strategy to replace the UPCs, though it relied on the same system of coordinated areas, which Mondragón implemented two years later.
    Called the Quadrants Plan, the system assigned responsibility for city blocks to individual police units in each coordinated area and used the police statistics system to track quadrant performance. Because UPCs had not existed in all 70 areas, levels of intensity of the use of monitoring statistics varied under Ebrard and Ortega. Under Mondragón, every coordinated area used the mapping information regularly. "Mondragón made use of CompStat information that already existed," Juan Salgado, scholar of police reform at the Center for Research and Teaching in Economics, said. "He invested in systems to make information available to the public and to more officers, but he didn't have to spend money on creating the data."
    According to Ebrard, the shift was necessary because the UPC program had achieved its goals. "When we started [in 2003], UPC meant the new kind of police that we wanted in the force, but by 2011, it didn't make sense to maintain two kinds of organizations within the police," Ebrard said. "We had replaced more than 35,000 people in this period, so that's why we finished in 2011: so we could have only one force with the same organization in the city." 
    According to Ebrard, he had considered using the Quadrants Plan earlier, but the system was unfeasible before the advent of the citywide surveillance camera network. "Why this change in 2009?" he said. "Because we expanded the video camera system so you can combine one system with another. If we didn't have the cameras, it would be impossible to have this system in the city, because you need a lot of police-140,000-on the force. It's impossible, very expensive. But if you use your presence with the cameras, you can do it."
    Ortega, however, said the system did not deviate significantly from the UPC program. "They changed the model into quadrants, but it's the same," he said. "It's the foundation of the UPCs."
    "The UPCs remain, structurally," Salgado said. "Check out the payroll of the SSPDF, and you'll see that the UPCs still exist-at least theoretically. However, the idea of creating them as the bulwark of police reform-including advancing the rights of officers and closer citizen-police relations-faded under Mondragón."




Innovations for Successful Societies makes its case studies and other publications available to all at no cost, under the guidelines of the Terms of Use listed below. The ISS Web repository is intended to serve as an idea bank, enabling practitioners and scholars to evaluate the pros and cons of different reform strategies and weigh the effects of context. ISS welcomes readers' feedback, including suggestions of additional topics and questions to be considered, corrections, and how case studies are being used: iss@princeton.edu. 

Terms of Use

Before using any materials downloaded from the Innovations for Successful Societies website, users must read and accept the terms on which we make these items available. The terms constitute a legal agreement between any person who seeks to use information available at www.princeton.edu/successfulsocieties and from Princeton University.
In downloading or otherwise employing this information, users indicate that
a. They understand that the materials downloaded from the website are protected under United States Copyright Law (Title 17, United States Code).  
b. They will use the material only for educational, scholarly, and other noncommercial purposes.   
c. They will not sell, transfer, assign, license, lease, or otherwise convey any portion of this information to any third party. (Republication or display on a third party's website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library.)
d. In all publications, presentations or other communications that incorporate or otherwise rely on information from this archive, they will acknowledge that such information was obtained through the Innovations for Successful Societies website.  A suggested citation format is as follows:
[Document author if listed], [Document title], Innovations for Successful Societies, Princeton University, accessed at http://www.princeton.edu/successfulsocieties on [date accessed on Web]
e. They understand that the quotes used in the case study reflect the interviewees' personal points of view.  Although all efforts have been made to ensure the accuracy of the information collected, Princeton University does not warrant the accuracy, completeness, timeliness, or other characteristics of any material available online. 
f. They acknowledge that the content and/or format of the archive and the site may be revised, updated or otherwise modified from time to time.   
g. They accept that access to and use of the archive are at their own risk. They shall not hold Princeton University liable for any loss or damages resulting from the use of information in the archive.  Princeton University assumes no liability for any errors or omissions with respect to the functioning of the archive. 
                           

Innovations for Successful Societies (ISS) is a joint program of Princeton University's Woodrow Wilson School of Public & International Affairs and the Bobst Center for Peace & Justice. The Woodrow Wilson School prepares students for careers in public service and supports scholarly research on policy and governance. The mission of the Bobst Center for Peace & Justice is to advance the cause of peace and justice through mutual understanding and respect for all ethnic traditions and religious faiths, both within countries and across national borders. 
References
1 Diane Davis, "The Giuliani Factor: Crime, Zero Tolerance Policing, and the Transformation of the Sphere in Downtown Mexico City."
2 Mascott Sánchez and María de los Ángeles, Seguridad Pública: Incidencia delictiva y sensación de seguridad, Centro de Estudios Sociales y Opinión Pública, 2002.
3 Reforming the Administration of Justice in Mexico, ed. Wayne A. Cornelius and David A. Shirk. South Bend, IN: University of Notre Dame Press, 2007, p. 9.
4 Business Mexico, "Marcelo Ebrard: Mexico City's  Security Secretary Talks about Crime, Corruption and Community Participation," October 1, 2002.
5 Angel Gustavo López-Montiel, "The Military, Political Power, and Police Relations in Mexico," Latin American Perspectives, Vol. 27, No. 2, March 2000, pp. 79-94; http://www.jstor.org/discover/10.2307/2634192.
6 For comparison, Mexico City had approximately 395 police per 100,000 population, and the New York City Police Department had approximately 499 sworn police officers per 100,000 residents, an all-time high for the department.
7 The country defined its moderate poverty line as the line below which people may be able to afford food but not education or health care. The extreme poverty line, below which people cannot afford sufficient food, was 672.27 pesos (US$70 in 2002) monthly in urban areas. Araceli Oretga Díaz, "Microstimulations for Poverty and Inequality in Mexico Using Parameters from a CGE Model"; http://s3.amazonaws.com/zanran_storage/www.colmex.mx/ContentPages/43641186.pdf
8 Nelson Arteaga Botello and Adrián López Rivera, "Everything in This Job is Money: Inside the Mexican Police," World Policy Journal 17(3)(2000): 61-70.
9 Valentine Anozie, Juhie Shinn, Katy Skarlatos, and Julio Urzua, "Reducing Incentives for Corruption in the Mexico City Police Force," International Workshop, Public Affairs 869, Spring 2004; http://www.lafollette.wisc.edu/publications/workshops/2003-2004/pa869/2004-mexico.pdf.
10 Mario Arroyo Juárez, "Evaluating the Zero Tolerance Strategy and Its Application in Mexico City," in Reforming the Administration of Justice in Mexico, p. 422.
11 Ibid., p. 430.
12 Geri Smith, "Mexico City's Battle to Beat Crime," Bloomberg Businessweek, August 19, 2003; http://www.businessweek.com/stories/2003-08-19/mexico-citys-battle-to-beat-crime.
13 Niels Uildriks, Mexico's Unrule of Law: Implementing Human Rights in Police and Judicial Reform under Democratization, Lexington Books, 2010, p. 109.
14 Diane Davis, "Zero Tolerance Policing, Stealth Real Estate Development and the Transformation of  Space: Evidence from Mexico City," Latin American Perspectives, November 20, 2012, p. 21; http://lap.sagepub.com/content/early/2012/11/19/0094582X12467761.full.pdf.
15 Smith, "Mexico City's Battle to Beat Crime."
16 Rodrigo Meneses Reyes, "Crime, Street Vendors and the Historical Downtown in Post-Giuliani Mexico City," International Journal of Criminology and Sociology, 2, 2013, p. 190.
17 Davis, "The Giuliani Factor."
18 "El centro, la zona más segura del DF: Ebrard," El Universal, April 16, 2004; http://www.eluniversal.com.mx/notas/217139.html.
19 For a detailed account of the incident, see Uildriks, Mexico's Unrule of Law.
20 "Informe de Actividades de la Secretaria de Seguridad Pública del Distrito Federal," 2005, p. 28.
21 "El centro, la zona más segura del DF: Ebrard," El Universal.
22 Valentine Anozie, Juhie Shinn, Katy Skarlatos, and Julio Urzua, "Reducing Incentives for Corruption in the Mexico City Police Force," International Workshop, Public Affairs 869, Spring 2004; http://www.lafollette.wisc.edu/publications/workshops/2003-2004/pa869/2004-mexico.pdf.
23 "Encuesta de Victimización y Eficacia Institucional," Centro de Estudios de Investigacíon y Docencia Económica, División de Estudios Jurídicos, 2012.
24 Nelia Tello, "Police reforms: The voice of police and residents in Mexico City," Policing and Society: An International Journal of Research and Policy, 22:1, 22-23.
25 Ibid.
26 Juan Salgado and Ernesto Treviño López, "Los Retos para la Reforma Policial a Nivel Local en México: Una Evaluación de las Unidades de Protección Ciudadana en el Distrito Federal," Fundar, Centro de Análisis e Investigación, 2008, p. 11.
27 Ibid., p. 67.
28 Ibid., p. 60.
29 Nathaniel Parish Flannery, "From Bullets to Bistros: The Mexico City Miracle," The Atlantic, February 3, 2013; http://www.theatlantic.com/international/archive/2013/02/from-bullets-to-bistros-the-mexico-city-miracle/272870/.
---------------

----------------------------------------

---------------

----------------------------------------




Michael Scharff
Innovations for Successful Societies





2

(c) 2012, Trustees of Princeton University  
Terms of use and citation format appear at the end of this document and at www.princeton.edu/successfulsocieties










ISS is a joint program of the Woodrow Wilson School of Public and International Affairs and the Bobst Center for Peace and Justice: www.princeton.edu/successfulsocieties. ISS invites readers to share feedback and information on how these cases are being used: iss@princeton.edu. 





Rachel Jackson
Innovations for Successful Societies





2

(c) 2014, Trustees of Princeton University
Terms of use and citation format appear at the end of this document and at www.princeton.edu/successfulsocieties
ISS invites readers to share feedback and information on how these case studies are being used: iss@princeton.edu 







ISS is a joint program of the Woodrow Wilson School of Public and International Affairs and the Bobst Center for Peace and Justice: www.princeton.edu/successfulsocieties




Rushda Majeed		Innovations for Successful Societies
Innovations for Successful Societies





4
(c) 2011, Trustees of Princeton University