Report from the Carter Center on the 1994 Mexican Elections Table of Contents SECTION PAGE The Council of Freely Elected Heads of Government 3 Executive Summary 4 PREFACE 6 INTRODUCTION 8 ELECTORAL REFORM ISSUES AND PROCESS 11 Part 1: Autonomy of the Federal Election Institute (IFE) 11 Views of the Actors 13 Part 2: Voter Registration and Voter Lists 15 The Padrsn 16 The Audits 17 A. External (McKinsey) Audit of the Padrsn 18 B. Party-Sponsored Audit of the Padrsn 20 Views of the Actors 20 Our Analysis 22 Part 3: The Problem of Media Bias 23 Views of the Actors 25 Part 4: Campaign Spending Limits 26 Views of the Actors 26 Part 5: Election Procedures 27 Views of the Actors 29 Part 6: National Observers 30 Views of the Actors 30 Part 7: International Observers 31 Views of the Actors 32 Part 8: Criminal Penalties; Creation of a Prosecutor's Office for Election Fraud. 32 Views of the Actors 33 ANALYSIS AND RECOMMENDATIONS 35 First-Order Issues 35 Second-Order Issues 36 Recommendations: 38 CONCLUSION 40 APPENDICES 41 Appendix 1: List of Meetings 42 Appendix 2: Biographies of Delegation Members 44 The Council of Freely Elected Heads of Government The Council of Freely Elected Heads of Government is an informal group of 24 current and former heads of government from throughout the Americas. The Council was established in November 1986 at a meeting chaired by former U.S. Presidents Jimmy Carter and Gerald Ford on "Reinforcing Democracy in the Americas" at The Carter Center. The Council's goals are to reinforce democracy in the Americas, promote multilateral efforts to resolve conflict in the hemisphere, and to advance regional economic cooperation. The Council has been a pioneer in mediating and observing elections. It has observed elections in Panama (1989, 1994), Nicaragua (1989-1990), the Dominican Republic (1990), Haiti (1987, 1990), Guyana (1990-1992) , and Paraguay (1993). The elections in Nicaragua and Haiti were the first free elections accepted by all parties in the nations' histories, and in Guyana, the first such elections in 28 years. The Council has worked since the elections to help consolidate democracy in Guyana and Nicaragua. In addition, the Council has a long-standing project in Mexico. In July 1992, four members of the Council sent representatives to witness the observation of elections in two states in Mexico; in November 1992, the Council invited a representative group of Mexicans to observe the U.S. presidential election; and in September 1993, a Council group visited Mexico to analyze the new Mexican electoral reforms. The Council is based at the Latin American and Caribbean Program of The Carter Center of Emory University. Dr. Robert Pastor, fellow at The Carter Center, is Executive Secretary of the Council; Dr. David Carroll is Associate Director; Dr. Jennifer McCoy, Senior Research Associate; and Ms. Harriette Martin, Administrative Assistant. COUNCIL OF FREELY ELECTED HEADS OF GOVERNMENT Jimmy Carter, former U.S. President, and Chairman of the Council George Price, former Prime Minister of Belize, Vice-Chairman Rafael Caldera, President of Venezuela (1969-1974, 1994-present) John Compton, Prime Minister of St. Lucia (1987-present) Luis Alberto Lacalle, President of Uruguay (1989-present) P.J. Patterson, Prime Minister of Jamaica (1992-present) Erskine Sandiford, Prime Minister of Barbados (1987-present) Jean-Bertrand Aristide, President of Haiti (1991-present) Razl Alfonsmn, former Argentine President (1983-1989) Nicolas Ardito-Barletta, former Panamanian President (1984-1985) Oscar Arias Sanchez, former Costa Rican President (1986-1990) Patricio Aylwin Azocar, former President of Chile (1990-1994) Fernando Belaunde Terry, former Peruvian President (1963-1968, 1980-1985) Rodrigo Carazo, former Costa Rican President (1978-1982) Vinicio Cerezo, former Guatemalan President (1986-1990) Joseph Clark, former Canadian Prime Minister (1979-1980) Gerald Ford, former U.S. President (1974-1977) Osvaldo Hurtado, former Ecuadorean President (1981-1984) Alfonso Lspez Michelsen, former Colombian President (1974-1978) Michael Manley, former Jamaican Prime Minister (1972-1980, 1988-1992) Carlos Andris Pirez, former Venezuelan President (1974-1979, 1989-1993) Julio Maria Sanguinetti, former Uruguayan President (1985-1989) Edward Seaga, former Jamaican Prime Minister (1980-1988) Pierre Trudeau, former Canadian Prime Minister (1968-1979) Executive Summary The national elections on August 21, 1994 will be an important milestone in Mexico's political opening. During the last four years, the Mexican Congress approved a number of important reforms to the electoral process. Yet the Mexican population remains highly skeptical about the integrity of the elections. Opinion polls show that nearly one-half of respondents expect fraud, and more than one half expect post-electoral violence. Substantial progress has been made in two historically controversial areas: the voter registration list and the vote count. The Federal Electoral Registry (RFE) has spent US$750 million to create a new voter registration list and issue new photo-identity voter credentials to ninety percent of the eligible voters. External audits commissioned by the Federal Electoral Institute and by eight of the nine political parties have declared the list to be reliable, but the PRD has not approved the audit results. Our preliminary analysis of the audit performed by the McKinsey consortium indicates that the voters list can adequately serve as the basis for a clean election, and that it represents a real achievement given the high degree of mobility of voters and the large size and diversity of Mexico. The opposition parties have identified two possible avenues for fraud related to the list. The McKinsey audit addresses one of those issues: the potential for padding the list with fantasmas or nonexistent persons is small. However, the audit does not directly address the second issue of the possible disfranchisement or exclusion of voters who believe themselves to be registered and qualified. Election day procedural changes, including private voting booths and transparent ballot boxes, as well as the accreditation of independent Mexican observers, improve the prospects for an accurate and verifiable vote count. The late invitation of foreign "visitors", however, severely limits the potential role of international observers as a vehicle for improving confidence in the process. An uneven playing field which limits the ability of all political parties to compete equitably remains of significant concern, especially regarding continued bias in media coverage, the high cost of advertising, campaign spending limits beyond the reach of any party except the PRI, and great disparities in financial resources. The greater independence of IFE's General Council is a marked improvement; however, the change in the General Council's composition only two months before the election leaves insufficient time for the new General Council to exercise control over the electoral machinery at the state and district levels and to ensure equitable conditions for a free election. The continued distrust of nearly half of the voters raises questions about whether the elections will be meaningful. If a plurality of voters believe their vote will not count and consequently do not vote their conscience, then even the most procedurally immaculate election will not reflect the will of the people. Accordingly, we recommend a set of actions that can help to raise confidence in these elections. The decision to invite foreign visitors came too late for the Council of Freely Elected Heads of Government to organize an observer mission similar to those we have fielded in other countries: a high-level delegation that has the capacity to develop over time ties of trust and respect with political leaders, to help overcome suspicion and miscommunication, and to make a comprehensive and global assessment of the entire electoral process. Instead, we will send a small group which will coordinate with others to report on the election day activities and the outcome. Recommendations to build confidence in the August 21 elections: 1. Hold additional public debates among the major presidential candidates. Of all the factors affecting public confidence, the May 12 debate appears to have had the most positive effect. 2. Leaders of the three major political parties should jointly film public service announcements on the need to vote and the voting process, and discuss ways to ensure the rapid dissemination of election results and a smooth transition to the newly-elected administration. 3. IFE should conduct an extensive civic education campaign to explain the procedures and safeguards, and the location of each casilla. 4. The final voters list should be posted by casilla at least ten days before the election to allow voters to confirm both their inclusion on the list and the location of their casilla. This should also help political parties and observers determine if there is any systematic exclusion of voters from the list. 5. Avoid exit polls, which are unreliable in a climate of suspicion and which will create a negative atmosphere if the voters feel they are being watched. Quick counts, however, are essential, but they must be well coordinated and the public needs to be informed of their significance. 6. Political parties should deploy party representatives to be present in as many casillas as possible; parties should coordinate to ensure that at least two parties are represented in each casilla. 7. National observers should undertake a mobile strategy on election day to cover as many polling sites as possible, paying special attention to polling sites where there are no opposition party representatives. The United Nations should coordinate international visitors. 8. The indelible ink should be guarded until election day and samples tested publicly to ensure that it has not been tampered with. Party representatives and observers should be especially vigilant to determine if voters' fingers are properly inspected and inked. 9. TV Azteca should match Televisa's donation of free time to the presidential candidates. Some of that time should be during primetime. 10. Political parties should make public reports of revenues and campaign expenditures before the election.PREFACE This report is the third on the Mexican Electoral Process prepared for the Council of Freely-Elected Heads of Government, which has been involved in Mexico during the last four years with issues related to the electoral process and to the North American Free Trade Agreement. Beginning in 1990, Mexican civic and political leaders have been invited to participate in the Council's election-monitoring missions in Haiti, Guyana, Paraguay and the United States. In the summer of 1992, the Council was invited by Mexican election-observer groups to send a small team to witness their observation of the elections in the states of Chihuahua and Michoacan and to report their findings. In September 1993, a small team representing three members of the Council visited Mexico to study the electoral reforms being discussed at that time in the Mexican Congress. That team's report to the Council, Electoral Reform in Mexico, was subsequently published and distributed widely in Mexico and the United States. It was also translated into Spanish and published in Este Pais in January 1994. The 1993 report was controversial, but we are gratified by the positive impact it seems to have had in Mexico. Many of the recommendations in that report were implemented, including the agreement to commission an external audit of the voter registration list, the increased autonomy of IFE from political party influence, greater access for national election observers, and the decision to accredit international visitors to observe the elections. In addition, national observer groups began to monitor IFE's decision-making and television reporting to discern whether there is bias in the coverage. This report seeks to put in context the latest round of electoral reforms and to assess the views of the major actors regarding the electoral rules and the playing field in the two months prior to the August 21 elections. It is based on a trip by a six-member team June 13-18, 1994 led by Dr. Jennifer McCoy, Senior Research Associate of the Carter Center and representative of Council member and former U.S. president Jimmy Carter, and including Canadian Senator Peter Stollery, representative of Council member and former Canadian prime minister Pierre Trudeau; Mr. Joaquin Daly, representative of Council member and former Peruvian president Fernando Belaunde Terry; Mr. Harry Neufeld, Canadian elections expert; Dr. Vikram Chand, a scholar of Mexico from Brown University; and Mr. Eric Bord, an attorney and consultant to the Carter Center. The report was drafted by Vikram Chand, Jennifer McCoy, Harry Neufeld, and Eric Bord, and reviewed and edited by Jennifer McCoy, Eric Bord, Joaquin Daly, Peter Stollery, and Robert Pastor. The delegation wishes to thank all of the persons who gave of their time and expertise in extensive meetings and telephone conversations in Mexico City. (A list of persons with whom the delegation met is provided in Appendix 1.) We also are grateful for the work of Harriette Martin, Administrative Assistant; and the following Carter Center interns who compiled extensive briefing material for the team: J.J. Gorsuch, Cynthia Hewitt, Joanne O'Connor, Colleen Shea, Hannah Temple, and Hewlett Summer Intern Alma Idiart. It is our aim to present an objective and impartial record of the changes in the Mexican electoral process for those interested persons in the international community. The following sections describe the most recent reforms of the Mexican electoral process and the views of the major political parties and national observer groups. The concluding sections give our analysis of the state of the Mexican elections and recommendations to further improve it. August 1, 1994 Atlanta, GA INTRODUCTION On August 21, 1994, Mexicans will elect a new President, 96 Senators and 500 Deputies to the National Chamber of Deputies. On the same day, six states and the Federal District (Mexico City) will hold state and local elections. These elections will undoubtedly be the most-watched in Mexican history and an important milestone in Mexico's political opening. Following the controversy and lack of credibility surrounding the 1988 elections, the government initiated two important sets of electoral reforms. The first was the 1990 election code which created a new election agency, the Federal Electoral Institute (IFE), a new voter registration list, and a photo-identity card. With mounting pressure from opposition parties and citizen groups, and questioned results in state elections held over the next three years, a second round of negotiations to further reform the process began in May 1993, and culminated in September with constitutional and electoral code changes. These are analyzed in our 1993 report. The traumatic events of the Chiapas rebellion on January 1, 1994 and the assassination of the PRI presidential candidate Luis Donaldo Colosio on March 23, 1994, spurred a new urgency toward a more open and credible election process. This led to political accords among the parties and further changes in the Constitution and election laws between March and June 1994. Credibility of the electoral process is the central issue for the 1994 elections. In June 1993, a poll of 1,400 persons conducted by Market Opinion Research Institute (MORI) showed that the population was evenly divided on the question "Do you think the 1994 elections will be clean or dirty" with 34% answering "clean", 34% responding "dirty", and 12% answering "so-so". Yet despite the changes in the rules and the tremendous effort and financial resources spent by the election authorities in the last year to ensure a clean process (US$750 million on the voter registration list and credentials), the Mexican population continues to doubt the credibility of the elections. Opinion polls taken in late May and early June 1994 indicate that almost one-half of the population expects fraud: a poll of 9,500 persons in twenty states conducted by the Civic Alliance, an umbrella organization of approximately 400 Mexican NGOs, and published June 30, showed that 47% expect fraud, while 28% do not. Likewise, weekly polls conducted by Miguel Basaqez of MORI from February to May indicated that 40-45% of respondents did not expect the vote to be respected, compared with 25-35% who did. A poll of 2,200 people published June 16, 1994 by the Mexico City newspaper Reforma showed greater confidence with 50% responding that they expect clean elections, yet 38% still responded that they do not. When asked who they expected to carry out fraud, 32% responded the government, 24.5% the PRI, and 13% the electoral authorities. (Civic Alliance, June 30). The one event that appeared to increase confidence was the unprecedented televised debate among the three main presidential candidates (PRI, PAN, PRD) on May 12. The MORI polls showed a jump in confidence in late May and early June, with opinion evenly divided between those who did and did not have confidence in the prospects for a clean election. Pollsters attribute the increase in confidence after the debates to the perception that the PRI was committed to a true competition for the first time, and that the PRI acknowledged the legitimacy of its competitors by pitting its candidate against the others. Similarly, a decline in confidence reported in MORI's June 26 poll may be a response to the attempted resignation of the Minister of the Interior (see below), while a surge in confidence in the July 10 poll is attributed to President Salinas' declaration that he will transfer power to whomever wins the election. The polls are significant in that they indicate that public opinion is not fixed, and that it may still be possible to increase confidence in the elections with well-publicized reforms and demonstrations of the government's commitment to a fair election.1 Even more alarming than the lack of confidence in the process is the high degree of uneasiness and fear of violence among the populace reflected in opinion polls. The June 30 Civic Alliance poll showed that in the event of electoral fraud, 30% expect violence with ungovernability and 35% expect violence with repression, while only 17% expect no violence. Similarly, in the June 16 Reforma poll, half the respondents said that post-electoral violence is very probable. Recent events underline the uncertainty and uneasiness surrounding the elections. The rejection of the peace accords by the Zapatista rebels on June 14 because of the lack of democratic guarantees left the Chiapas problem dangerously unsettled. The June 15 resignation of the Peace Commissioner, former Mexico City mayor Manuel Camacho Solis, with an acrimonious letter criticizing PRI candidate Ernesto Zedillo, raised more speculation about alternative scenarios for an interim government in the wake of an unclear election. Then, on June 24, Minister of Interior Jorge Carpizo, submitted his resignation to President Carlos Salinas de Gortari out of concern that some partisan elements were threatening the integrity of the elections. President Salinas convinced Carpizo to withdraw his resignation after lengthy conversations between the two and calls from all three major parties for Carpizo to stay. In an ironic twist, this event magnified the important issue of IFE's independence; despite the work of the parties and government to increase the autonomy of IFE, it was they who turned to the government minister historically in charge of elections in order to ensure the integrity of the process. In response to the tense climate, a diverse group of fifty intellectuals and political leaders organized themselves out of concern that the results of the election may be questioned. Calling themselves the Grupo de San Angel, on June 29, they issued a public declaration, "La Hora de la Democracia," in which they committed themselves to work toward clean elections and to develop an agenda to promote a post-electoral transition to democratic modernization. The polls are significant in that they indicate that public opinion is not fixed, and that it may still be possible to increase confidence in the elections with well-publicized reforms and demonstrations of the government's commitment to a fair election. ELECTORAL REFORM ISSUES AND PROCESS From March through June 1994, the Mexican Congress, with the support of the PRI, PAN and part of the PRD, approved a number of important reforms to the Constitution and election laws (COFIPE), and IFE promulgated several significant administrative changes. The reforms to the electoral code: enhance the autonomy of the Federal Election Institute (IFE); create a special prosecutor's office to investigate and punish violations of election law; impose stiff criminal penalties for various sorts of election fraud; improve access for national observers; and invite and accredit international visitors/observers. These reforms represent an important advance beyond the election reforms approved by the Mexican Congress between August 30 and September 15, 1993. The new IFE guidelines address: equitable access to the media; campaign spending limits; and new election day procedures. In addition, IFE sponsored two audits of the voter registration list (Padrsn), one by a consortium of external professional organizations, and the other in collaboration with eight of the nine political parties (the PRD chose not to participate). This section will discuss each of these issues, including how they are perceived by Mexico's major political parties and national observer groups. Part 1: Autonomy of the Federal Election Institute (IFE) IFE is the organization charged with the management of federal elections in Mexico. IFE is run by a director-general nominated by the Minister of Interior and approved by a 2/3 vote of the General Council. The General Council is responsible for setting policy for IFE, and it is chaired by the Minister of the Interior (Gobernacisn). IFE is divided administratively among six specialized organs responsible for compiling the voter registration list, organizing elections, supervising the Professional Electoral Service (SPE), monitoring party finances and access to the media, and approving coalitions among parties. IFE's six specialized organs are staffed by members of the Professional Electoral Service (SPE). The national structure of IFE is replicated at the state and district levels with local councils (Consejos Locales) ruling on federal elections in Mexico's 31 states and the Federal District, and district Councils (Consejos Distritales) in Mexico's 300 single-member districts ruling on election to the Chamber of Deputies. In addition, each of IFE's six administrative organs has field offices at the state levels in the local executive board and at the district level in the district executive boards. Prior to the latest round of reforms, the General Council consisted of 21 members with the right to vote, including one representative each from the two largest parties in the Chamber of Deputies, one representative each from the two largest parties in the Senate, six Magistrate Councilors not linked to any party, and representatives of political parties allocated according to a formula reflecting their relative strength in the last federal elections for deputies. The President of the General Council was permitted to vote but in practice did so only to break a tie. The Magistrate Councilors - required by law to be lawyers or judges - were appointed by the Mexican President and approved by a 2/3 vote of the Chamber of Deputies. In 1993, the government had one seat on the Council; the PRI had six; the PAN had three; and the PRD two. In addition, each of three small parties had one: the Popular Socialist Party (PPS), the Authentic Party of the Mexican Revolution (PARM), and the Cardenista Front for National Reconstruction (PFCRN). Technically, the PRI and government together controlled only seven seats in the General Council under the old system. Yet, the PAN and PRD believed that the PRI/government could actually secure a majority of the votes or even a 2/3 majority through the votes of some or all of the six Magistrate Councilors and the three small parties. Our own study of voting patterns in the General Council revealed that the Magistrate Councilors would normally side with the PRI/government on important issues; the small parties tended to vote similarly although not on every important issue. (see our 1993 Report, pp. 23-28). The latest round of reforms seeks to address this perception of bias among the Magistrate Councilors and strengthen the autonomy of the IFE. The Magistrate Councilors were replaced in early June 1994 by six Citizen Councilors appointed through consensus by the three main political parties, rather than at the behest of the President, and approved by a 2/3 vote of the Chamber of Deputies. The requirement that they be lawyers has been dropped, thus widening the potential pool of candidates and reducing their ties to the judicial branch of government. The Citizen Councilors must not have held any position of leadership in a political party in the three years prior to their designation. The new Citizen Councilors consist of a former columnist, three academics respected for their independence, a former public official, and a legal adviser to the corporate sector. New Citizen Councilors Santiago Creel Miranda, Academic Miguel Angel Granados Chapa, Journalist Josi Augustmn Ortmz Pinchetti, Lawyer and Political Analyst Ricardo Pozas Horcasitas, Sociologist and Political Researcher Josi Woldenberg Karakowsky, Sociologist and Political Analyst Fernando Zertuche Mzqoz, Academic The new reforms also give political parties the right to participate in General Council deliberations but do not allow party representatives to vote. The objective of this reform is to depoliticize General Council decision-making processes by increasing the relative weight of the Citizen Councilors who now control a majority of the votes on the Council. The General Council will henceforth consist of only 11 voting members including the six Citizen Councilors, one representative each from the two largest parties in the Chamber of Deputies, one representative each from the two largest parties in the Senate, and the President. The President will have the right to vote in all Council decisions, but may not break a tie. However, the President has announced that he will not exercise his vote in order to enhance his authority as an impartial arbiter of Council proceedings. The new President of the General Council, Minister of Interior Jorge Carpizo, is well known for his personal integrity and concern for human rights; he served as President of Mexico's National Human Rights Commission prior to his appointment as Minister of the Interior and President of the General Council of IFE. The initial deadline for submitting the names of such individuals along with appropriate evidence was extended from March 5, 1994 to June 14, 1994. However, the General Council President indicated on June 18 that additional names could still be submitted after the June 14 deadline. By June 14 approximately 170 citizen councilors at the district and state levels had been removed out of 1,992 councilors, or about 8.5% or the total. In addition, approximately 234 members of the Professional Electoral Service (SPE) at the state and district levels were removed at the request of opposition political parties. Polling station officials for election day in the approximately 97,000 casillas have already been chosen by two successive lotteries to ensure their objectivity. In the first lottery, 15% of all citizens from each electoral section were selected at random to participate in a training course given by IFE to serve as polling station officials. The lottery was held in the central headquarters of IFE in full view of the press. A second lottery was held after the training course and graduates of the course whose last name began with a randomly chosen alphabetical letter were selected as officials for their sections. Rank within the polling station was then determined by level of education. In addition, the Director-General of IFE, with the approval of 2/3 of the General Council, appointed Dr. Juan Molinar, a well-regarded independent critic and scholar of the Mexican electoral system, to head the Office of Political Parties and Prerogatives which deals with the public financing of political parties, media access, and the registration of coalitions. Views of the Actors The PRI, PRD, and PAN all see the changes in the composition of the General Council as very positive steps. The PRI believes that the appointment of the new Citizen Councilors will add credibility to the electoral process, thereby reducing the risk of a disputed outcome later on. The two largest opposition parties - the PAN and the PRD - see the new Citizen Councilors as objective, independent, and open to their concerns. The Citizen Councilors themselves are aware that their greatest strength is their moral authority. They have already used their power of moral persuasion to criticize the electronic media for covering the candidates in a partial and biased fashion (see discussion below in Part 3), and there are preliminary indications that this criticism is already having an impact. The Citizen Councilors are also conscious of the possibility that the citizenry may turn to them to resolve any post-election crisis because of their moral credibility and nonpartisan image. Both opposition parties and the Citizen Councilors, however, are concerned about the problem of unrealistic expectations regarding the ability of the Citizen Councilors to monitor and regulate the performance of the electoral bureaucracy. They point out that while all the Magistrate Councilors at the General Council have been replaced, the rest of IFE as an institution remains largely intact. Less than ten percent of the combined Citizen Councilors and electoral officials at state and district levels have been removed, and five of the six executive organs of IFE remain under the control of officials appointed under the old rules, that is directly by the director-general without the approval of the General Council. The presence of the new Citizen Councilors at the apex of IFE thus does not guarantee a transformation of IFE as a whole. In addition, the new Citizen Councilors were appointed just two months before the elections, thereby limiting their impact to the last phase of the electoral process. The PRI, PRD, and PAN all see the changes in the composition of the General Council as very positive steps. The national election observer groups, particularly the Civic Alliance, also view the changes in IFE as a positive step. But the Civic Alliance criticizes opposition parties, particularly the PRD, for not doing enough to identify biased Citizen Councilors and officials at the state and district levels. In many cases, the PRD submitted only the names of suspect councilors and officials without any accompanying evidence. The Civic Alliance also claims that the PRD and PAN had enough time to challenge corrupt officials, particularly since the deadline was extended through June 14, 1994. The PAN and PRD feel that it was very difficult to collect evidence against suspect councilors and officials and that it would have been better to dismiss all 1,992 Citizen Councilors at the state and district levels en masse. The government responds that this would have been impractical so late in the electoral process.Part 2: Voter Registration and Voter Lists The Government of Mexico chose to create a computerized Register of Voters (Padrsn) at the time that a new electoral roll was prepared in 1991. Subsequently, the General Council of the IFE decided to enhance the registration process by issuing each registrant an identity card which incorporates the voter's signature, photograph, and fingerprint. The photo-identity card features nine security measures and a magnetic strip which could be employed for electronic voting in the future.2Keeping a register of voters current in Mexico presents major challenges. Three measurements are traditionally used to evaluate the quality of a voters list: Completeness Currency Accuracy Completeness addresses the question "Is everyone on the list that should be on the list?". Not all who are qualified need necessarily register. In most Western democracies it is generally accepted that up to 10% of the eligible population does not wish to register and therefore does not appear on the list. The Mexican government estimates there are currently 50 million eligible voters, representing an average increase of 1.6 million eligible voters per year since 1991. The IFE estimates that 95% of eligible citizens actually registered to vote (47.5 million out of 50 million). Of these 47.5 million, slightly more than 95% (45 million) had collected their voter credentials and will appear on the August 21 voter list. This represents a completeness coverage of 90.25%. Currency is a measure of the relative decay over time of register information following its initial preparation, offset by maintenance efforts. Knowing the currency answers the question, "Is the list up to date?". The Registro Federal Electoral (RFE) estimates a national mobility rate of approximately 6% annually, an adult death rate of approximately .06% per year, and approximately 8% of the workforce employed as migrant labor (however many have a permanent address they regard as home even though they may reside there only periodically). Given these factors, keeping a register of voters current in Mexico presents major challenges. The RFE estimates that since the photo-identity program began in January 1993, it has received 13 million new applications for registration, processed 2.4 million changes of address, made 1.4 million data corrections, and canceled 5.1 million outdated registrations, mainly due to unreported changes of address or deaths. The comprehensive audits of the list (discussed further below) provide indications of the list's currency. In most Western democracies a currency level of 85% is considered to be a high level of maintenance if it is based on a voluntary system of change-of-address updates. Accuracy measures both the substantive and cosmetic errors on the list and addresses the question, "What percentage of the names and addresses on the list are correct?" Substantive errors include the registration of voters who do not exist, listing of voters at nonexistent addresses, duplicate registrations, and assignment of voters to the wrong voting section. Cosmetic errors, which would have no real effect on the electoral process, include a misspelled name or address, and the misrecording of gender or age information. Though 100% accuracy is the goal, experience indicates that a 97% accuracy rate is the best that can reasonably be expected because of non-intentional human errors. A high rate of accuracy would, therefore, still expect cosmetic inaccuracies in 2% of the entries and substantive inaccuracies in 1% of the list. However, if an organized campaign of collecting improper registrations at the source is mounted, for motives of political bias or some other reason, the rate of substantive error can be expected to increase markedly. The Padrsn Our 1993 report provides an overview of the creation of the Padrsn being prepared in anticipation of the August 21, 1994 elections. Since that report, a number of significant operational improvements and legal reforms have been implemented. They include: deletion of 5.1 million outdated registrations and photo-identity cards that had not been picked up at the 7,000 registration offices across the country; identification and deletion of approximately 100,000 duplicate registrations, 20,000 duplicated photo-identity cards and the names of 30,000 persons sentenced for crimes. Many of these records were identified through the use of computer searches for similar sounding names and ranges of birth dates; enactment of laws which make it a criminal offense to be in possession of more than one photo-identity card; legal access for representatives of political parties to the RFE's archives of registration source documents and photographs, housed in seventeen locations throughout the country; agreement to attach an "addendum", if necessary, to each voting precinct list showing the names of persons who appeal to the Federal Electoral Tribunal and are ruled eligible to be included on the list of eligible voters after the final list (Listado Nominal Definitiva) is produced. Some 90,000 such administrative requests made by the Tribunal had been processed by the RFE at the time of the delegation's visit; the funding of political parties for the purpose of undertaking a national audit of the Padrsn and its sources; and establishment of an Electoral Roll Technical Council, made up of ten politically impartial professionals and experts, that has responsibility to oversee the comprehensive multifaceted external audits of the Padrsn and its preparation by independent national and international firms. Notwithstanding the efforts and expense (US$750 million) to which the IFE has gone to assure the integrity of the registration process and the vote, there remains widespread mistrust among the Mexican populace regarding the electoral process in general (discussed in the Introduction of this report), and specific questions of the Padrsn's reliability among the opposition parties. Throughout the surrounding controversy, RFE officials have been steadfast in their defense of the Padrsn and the professionalism they have applied to its creation and maintenance. They point to the positive results of audits taken in a Selectivity Study in 1991, a National Verification in 1992, and 36 separate and methodologically diverse studies taken since the photo-identity card issuance program began in November, 1992. The Mexican voters roll is not perfect, they admit. However, they claim to have studiously improved their processes as a result of suggestions made during each of these investigations and argue that they have produced a very reliable list in comparison to that of any other nation. On July 20, 1994, based on 38 separate studies including the two audits discussed below, the IFE General Council approved the Padrsn and nominal lists. PRD Senator Porfirio Muqoz Ledo was the only one to vote against acceptance. (El Financiero, International Edition, July 25-31, 1994). The Audits As a confidence-building measure, the IFE General Council agreed to fund two separate audit processes at a cost of over US$11 million. One audit was conducted by a consortium of national and international professional organizations. The other was commissioned by the IFE along with eight of the nine political parties (the PRD did not participate). The audits were concluded on June 30, 1994, and the results compiled and delivered to the IFE shortly thereafter. Results and conclusions described in this section are based upon summaries issued by the IFE in Press Releases dated July 8, 1994 and July 10, 1994 and a preliminary review of the external (McKinsey) audit. We did not receive the political party audit in time to analyze its conclusions. Neither did we receive the audits of the PRD, although we repeatedly requested them in order to examine their criticisms of the Padrsn. A. External (McKinsey) Audit of the Padrsn One of the audits was performed by a consortium of eight (three international, five national) Mexico based companies and involved investigations of computing processes, adherence to legal procedures, a sample check of 64,000 data base records and source documents against interviews of voters in their homes to determine currency and accuracy, and a resource usage evaluation that compared the number of photo-identity cards manufactured and issued to the amount of material used. This audit process was overseen by a Technical Council of ten politically independent technicians (seven with PhDs) who were responsible to the General Council of the RFE for: determining the reliability of the Padrsn (registered voters) and Listado Nominal Definitiva (registered voters with a photo-identity card); identifying the causes of failures or inconsistencies that were found; and, making suggestions to the IFE General Council of corrective measures. The companies involved in the audit included: Berumen y Asociados Bufete de Consultorma Actuarial Burs de Investigacisn de Mercados Grupo de Asesores Unidos S.C. A.C. Nielsen McKinsey & Co. Salles, Saenz y Co. Systemhouse de Mixico McKinsey & Co. acted as the coordinator of the other seven firms and was responsible for integrating the audit results and producing the final report. It is significant that the audit went much further than simply testing the quality of the list; it was designed to investigate areas of potential mismanagement and questionable or illegal practices used in the process of list creation. The principle findings of the external audit were: The Padrsn is 97.42% reliable; Processes utilized in producing and updating the Padrsn and photo-identity cards were within all legal norms; No extraneous or additional information is present in the information systems; and, The amount of material used in creating the photo-identity cards reasonably corresponds with the number of cards issued. According to the July 8 IFE press release, the Technical Council concluded that the methodology used by the contracted firms complied with the criteria and established norms that were promised in their bid. The Technical Council further certified that the audit was performed in accordance with scientific standards and generally accepted methods, and that the results of the audit should be considered valid. The margin of error was 0.5% with a 95% level of confidence at the national level.The external auditors concluded that the size, structure, and composition of the small level of inconsistencies should reassure those concerned that the Padrsn is reliable and that there is nothing to suggest bias or systematic violation of normal procedures. The report of the audit indicates a reliability level for the Padrsn of 97.42%. This figure attempts to combine elements of accuracy and currency, and includes not only those cases where all of the information coincided between the data base, source document, and the citizens surveyed (78.70% of the cases), but also the following categories: changes of residence where the person was known at the original address and found at a new address by the surveyor (4.96%); changes of residence where the person was known at the original address but not found in the same vicinity by the surveyor (4.56%); citizen reported the photo-identity card lost or stolen or the citizen had moved temporarily (4.7%); the citizen had died (0.29%); and the citizen had the credential but was not on the preliminary voters list of April 30, 1994 used to draw the sample (4.25%) The RFE has made assurances that inconsistencies in this last category were remedied in the final updating of the voters list. Furthermore, to the extent that there were inconsistencies, they were dispersed throughout the country and not geographically centralized. The cases that were counted as inconsistencies in this audit (2.37% of the total), then, were only those cases in which either the address was not found or the citizen was unknown at the address (1.9%); citizens without credentials but listed as having them (0.3%); citizens who had returned their credentials because of errors (0.07%); and cases where the data coincided with the citizen interview, but the source material was missing (0.1%). The external auditors concluded that the size, structure, and composition of the small level of inconsistencies should reassure those concerned that the Padrsn is reliable and that there is nothing to suggest bias or systematic violation of normal procedures. The audit went on to conclude that the software programs, information processes, and the data bases contain no extraneous programming or information, and include only that which is proper and necessary for the operation of those systems. In addition, the computerized data at the regional computing centers and the data given to the parties correspond. With respect to the updating that was performed by the RFE, the audit indicates that there were no irregularities and that all legal norms were followed. In addition, the material used in processing the photo-identity cards reasonably corresponds to the number that were printed and distributed. The auditors also concluded that the actions and activities of the RFE complied fully with all legal requirements and all regulations and orders issued by the General Council of the IFE and the National Commission of Oversight, a council comprised of representatives from the nine political parties. Some questions were raised regarding the vulnerability of the information systems to tampering. Recommendations for immediate additional security measures were made, and, according to press releases, have been implemented by the IFE. Those suggestions include improved physical security and hazard protection at computing centers and more rigid control over access to the data base through the increased use of passwords. B. Party-Sponsored Audit of the Padrsn This audit was commissioned by all of the political parties except for the PRD. Its results were accepted by seven of the nine parties, with the PDM abstaining3 and the PRD voting against approval. Its objective was to determine (a) the level of correspondence between applications, receipts and data base records; and (b) the existence of citizens at their reported address. The methodology employed included a sample of 82,405 citizens throughout the 31 states and the federal district. The margin of error was limited to 0.63% with a 98% confidence level at the national level. These measures of accuracy and currency should provide a reliable snapshot from which to extrapolate to the entire population. The results of the party-sponsored audit as reported in the IFE press release of July 8 appear to be consistent with the conclusions of the external audit: the reliability of the Padrsn was measured at between 96.08% and 97.62%. However, we are unable to provide our own evaluation of the party-sponsored audit because we had not received the audit itself at the time of this writing. According to IFE, the eight sponsoring parties agreed that the level of reliability of the Padrsn is a consequence of the timely corrective measures that were implemented by the RFE, specifically the removal of duplications, searches for similar sounding names, changes of address and changes of citizenship. The PRD representative before the National Oversight Commission, Josi Barberan Falcsn, asked that the PRD be furnished with the results of the audit so that they could be reviewed by the PRD. According to the RFE, the PRD was provided the results of the audit, but denied the database for the audit on grounds of confidentiality. Views of the Actors The PRD refused to be part of the party-commissioned audit. It continues to be extremely skeptical of all of the activities of the RFE and claims that it has been given "no answers" and has "great doubts" about the problems it has identified with voter registration. The PRD claims that there is an error rate of 20% in the Padrsn. Based on a survey in 45 electoral districts in four different states, the PRD found that 9.5% of the names on the lists were "fantasmas" or "ghosts" (nonexistent persons), while 11% of "real" citizens were excluded from the lists. ("RFE Responds to PRD Arguments in May 30, 1994 Debate on the Padrsn" in Report of IFE General Council meetings published June 10, 1994). The RFE vigorously denies the accuracy of the PRD surveys and declares that it was able to find more than 90% of the persons associated with ghost names within a three day period. The RFE further suggests that the PRD is prone to extrapolate its figures on a basis that has no statistical merit and lacks the application of rigorous methodology. The RFE is concerned that the PRD is trying to discredit the entire electoral process on the basis of isolated registration errors. The PAN and PRD are also concerned that the government may alter the final voter registration list (Listado Nominal Definitiva) on the eve of the elections by reducing the number of voters in areas where the opposition has historically done well. Three agreements attempt to address this concern. First, the General Council voted to post the final list in the municipal city halls between August 10 and August 20 so that citizens can check it. (El Financiero, 21 July, 1994). Ten days should be adequate time for this purpose. Second, on election day representatives of the political parties at the district level will each select at random their copies of the Listado Nominal Definitiva, with the final remaining copy to be that which is used by the officials at each voting site. (IFE Press Release, July 8, 1994). This should ensure that a unique, tampered list is not provided to polling station officials, but it will still require that the political parties compare the official list with their own randomly chosen copy, and with the Listado Nominal Definitiva provided to the parties before election day. Third, the Citizen Councilors proposed at the July 20 General Council meeting to conduct a sample analysis of the final registration list used on election day to ensure that it is identical to the lists received by the political parties. The PRD is concerned that the government may possess a secret software program that might be activated to alter the voters list shortly before the elections. The government strongly denies the existence of a secret software program, and the external audit reportedly found no extraneous programming or information. One international group, the United Nations Electoral Assistance Program, claims that one way to guard against alterations in the final voter registration list is to introduce a system of tendered balloting. Under such a system, voters could vote with their credentials even if their names were not on the list. These votes would then be sent to IFE, which would verify the validity of each of the ballots in cooperation with the parties. This would enable IFE and the observer groups to quantify the number of citizens whose names should have appeared on the final registration list but did not; verify if there was a pattern of bias in the exclusions in favor of any given party; and allow those excluded to vote. The government, the Citizen Councilors, and parties, however, are uncomfortable with the system of tendered balloting because it permits people to vote who are not on the final voter registration list, a traditional avenue of fraud in Mexican elections. Our Analysis The Carter Center commissioned a voters list expert and a statistician to review the information provided in the external (McKinsey) audit. Our preliminary conclusion is that the Padrsn provides a good basis for a free election, and is a real achievement in light of the challenges faced by the RFE in compiling a list in a country as large and diverse as Mexico. Nevertheless, we find the conclusion that the Padrsn is 97.5% reliable to the most optimistic interpretation of the data, and somewhat confusing. This is due to the fact that the criteria measured by the audit blends issues of accuracy and currency (as defined above). The audit seeks to answer the following questions: a) do the addresses on the list exist and are they located in the correct electoral section; b) do the citizens on the list exist and are they in the correct electoral section; c) what percentage of voters have received their photo credential; and d) what is the level of consistency between the source documents and the data bases? Of these, items (a), (b), and (d) answer the question of how accurate the list is. Yet, the inconsistencies identified in the audit combine issues of currency (changes of residence and deaths) with issues of accuracy (does the house or person exist). The result is a single percentage of "consistency" or reliability that is difficult to interpret. Our preliminary conclusion is that the Padrsn provides a good basis for a free election, and is a real achievement in light of the challenges faced by the RFE in compiling a list in a country as large and diverse as Mexico. The implications of these findings for discerning a pattern of bias or potential avenues of fraud in the list are unclear. On the one hand, the consistency rate of 78.7% indicated in the audit (where all of the data matches between the citizen interviewed, the source documents, and the computer database) is actually quite good since this reflects all those cases which met both of the criteria of accuracy and currency. In addition, the potential for padding the list with invented persons (fantasmas) appears to be quite small: only 1.9% of registered voters were not known at the address listed, some of which could potentially be nonexistent persons. On the other hand, several important questions are apparently not answered by the data provided in the audit summary. First, among the cases of changed addresses, how many people might be registered at more than one address? (The RFE had an extensive system to check for duplicate registrations by comparing similar sounding names within a range of birthdates to combat this potential problem.) Second, how many people are excluded from the list involuntarily? This is an issue of coverage which is not addressed in the external (McKinsey) audit. If 90% of the eligible voters are actually on the list, were the remaining 10% disenfranchised involuntarily or did they simply fail to register due to lack of interest? To answer this question would require a survey that is drawn from the pool of eligible voters and then checks them against the list (people-to-list audit), in addition to the list-to-people audit that was carried out by the McKinsey group. A people-to-list audit would attempt to discern if there was a systematic bias either by geographic location or political preference that could constitute one form of fraud: the deliberate disfranchisement of people by excluding them from the list. Other safeguards must protect against these two primary forms of fraud in a voter's list. The available safeguards in this context are vigilant use of indelible ink to prevent multiple voting, and the review of the final voter's list to ensure that all registered voters are in fact on the list and permitted to vote. The posting of the lists in municipalities ten days before the election therefore provides an important opportunity for voters to ensure that they are correctly listed, or to submit appeals which could later be examined to determine if there was a biased disfranchisement of voters. Finally, we should note that, according to Dr. Carlos Almada, Director of RFE, it is not an electoral crime to vote at one's old address if that is the address on the credential and the list. Almada explained that the political parties agreed that this would be permissible because of the high mobility level of urban Mexican adults, the significant number of migrant workers, and a tendency in some sectors for voters to consider their ancestral village as home even though they may reside elsewhere. Part 3: The Problem of Media Bias The 1993 electoral reforms directed IFE to arrange for radio and television time for each of the parties to be paid for by the state and additional time to be paid for by the parties, but did nothing to address the problem of bias by the country's television stations in favor of the official party. On January 27, 1994, all the country's political parties and presidential candidates reached an Agreement for Peace, Democracy, and Justice in which they agreed that equal access to the media was a necessary precondition for fair elections. A May 19 study of news coverage from January - April 1994 on 24 Horas and Hechos by the Mexican Academy of Human Rights, one of the members of Civic Alliance, found that the PRI enjoyed a 3:1 advantage in total air time compared with the two strongest opposition parties, the PAN and the PRD. The coverage of presidential candidates was even more skewed with the PRI candidate receiving 3.5 times more coverage than the PAN candidate and 6 times more coverage than the PRD candidate. The Civic Alliance plans to continue monitoring newscasts for bias and report its findings to the citizenry on a regular basis. A second report based on content analysis of television coverage during June 1994 found a slight improvement in total air time given to presidential candidates, but it was still skewed toward PRI, particularly on Televisa where Zedillo received 36%, Cardenas 11%, and Fernandez 8% of the coverage on 24 Horas. The new General Council has taken three steps to remedy the issue of media bias by radio and television stations. First, the General Council has used its moral authority to formally exhort radio and television stations to provide more equitable coverage of all political parties, particularly in their newscasts. On June 18, it reminded the owners of Mexico's radio and television stations that the right to information and the free expression of ideas are part of the Universal Declaration of Human Rights of the United Nations to which Mexico is a signatory. The General Council has also communicated the concern of various citizen organizations and parties about the lack of fairness and objectivity with regard to the coverage of the presidential candidates. The June 1994 study by the Mexican Academy of Human Rights found that when value judgments about candidates were injected into stories on Televisa's 24 Horas, PRI candidate Zedillo was praised at a disproportionate rate of 12:1 over his closest rivals. Second, the General Council established a commission to monitor the country's radio and television news programs for biases in coverage and report back to the General Council by June 30. The commission will continue to provide regular reports on news coverage by the electronic media to each ordinary session of the General Council, and the General Council will continue to formulate recommendations to the media to improve the fairness of its coverage of candidates and parties. The first report, based on media monitoring from June 22-28, 1994 showed that the PRI received nearly as much radio and television coverage as the eight opposition parties combined (Miami Herald, July 6, 1994), confirming the results of the Mexican Academy of Human Rights. Third, the General Council plans to provide approximately two million U.S. dollars to political parties to purchase time on radio stations. A portion of this money will be allocated according to the electoral strength of each party as reflected in the previous federal elections for deputies; the rest will be allocated equally among all parties. According to press reports of July 8, 1994, Televisa, which has approximately 90% of the viewing audience, announced it would invite all nine presidential candidates to make three 15-minute political broadcasts which would be shown repeatedly up to a week before the elections. The programs will be shown at different times to ensure they reach all audiences. This announcement follows a meeting between Civic Alliance and President Salinas on Friday, July 8 in which Civic Alliance accused Televisa and the smaller Television Azteca of slanting their coverage in favor of the PRI candidate. Television Azteca has not yet replied to the charges. Both the PAN and the PRD welcomed the Televisa move and plan to record their programs shortly, although the PRD said that it did not go far enough in ensuring completely fair coverage. In addition, the PRI, PAN, and PRD reached an agreement that the government would suspend publicity for its two antipoverty programs, the National Solidarity Program (PRONASOL) and PROCAMPO, during the last three weeks of the campaign. Views of the Actors The PAN and the PRD support the steps that the General Council has taken to encourage equitable access to the electronic media though they note that the General Council's recommendations are not legally binding on the owners of Mexico's radio and television stations. The opposition parties are also pleased with the agreement to ban all publicity for PRONASOL and PROCAMPO during the last three weeks of the campaign because of the risk that the government might use such publicity for partisan ends. The PAN and PRD remain deeply concerned about the problem of bias in the electronic media since the overwhelming majority of Mexicans receive their information about the campaign from television newscasts. The PAN and PRD feel that Mexico's two most important TV news programs, 24 Horas of the Televisa Network and Hechos of the Azteca Television Network, are biased in favor of the PRI. The PAN and the PRD argue that the owners of Mexico radio and TV stations are more likely to support the ruling PRI because they receive their licenses from the government. In addition, the PAN argues that the high rates for purchasing television slots in prime time (approximately US$120,000 per minute) favor the PRI, which has access to more financial resources than the PAN and the PRD. The PAN would like the PRI to disclose publicly how much it has spent on advertising in the electronic media to date. The PAN did concede that there has been a slight improvement in the quality of coverage of the campaign, particularly radio. The PRI states that it cannot do anything about the bias in the electronic media because they are in private hands. The PRI also claims that it is the most important political force in the country and hence the most newsworthy. The PRI stated that is had not released any figures on advertising in the electronic media as of June 15 because it began purchasing spots on television only recently, and in addition notes that it is under no legal obligation to release its campaign expenditures until after the election. The government for its part argues that it cannot regulate the content of newscasts without seriously compromising the principle of free expression. Part 4: Campaign Spending Limits The 1993 electoral reforms provided for campaign spending limits for the first time in Mexican electoral history but left it up to the General Council to specify the limits as well as the formula and criteria used to derive the limits. Prior to the change in composition of the General Council, it set a limit of US$42 million for the presidential race for each political party. To arrive at this figure the General Council set a monetary value on each vote multiplied by the number of registered voters, and adjusted for both inflation and the duration of the campaign. Campaign spending limits for Mexico's senate and federal deputy races were approved by the local and district councils respectively. The formulas used were the same as the one used for the presidential race except that they take into account two additional factors: geographical expanse and population density. The limits for senate races range from a high of approximately nine million U.S. dollars in the State of Mexico to a low of approximately US$426,000 in the State of Baja California Sur. The limits for federal deputy races range from a high of approximately US$653,000 in District VIII of the State of Mexico to a low of approximately US$45,000 in District XIX, also in the State of Mexico. Views of the Actors The PAN and the PRD feel that the limit of US$42 million per presidential campaign is much too high. The PAN has stated that the official limit is eight times more than what it plans to spend on the presidential campaign. The opposition parties feel that the high campaign spending limits give the ruling PRI an advantage since it has greater access to resources. The PAN and PRD are also worried that money used for the government's social programs will be used to buy the vote for the official party, particularly in rural areas. The PAN claims that the government is planning to spend around US$4 billion in the two months preceding the elections through PRONASOL and PROCAMPO. The government argues that it cannot halt investment in public programs merely because of the elections. It claims that the very fact that it is channeling funds into poverty-stricken areas is a sign of its democratic responsiveness and that all democratic governments use social programs to garner political support. The government also claims that with a secret ballot, just because a citizen happens to benefit from a public program does not automatically mean that the same citizen is obliged to vote for the ruling party. Finally, the government points out that the latest reforms establish severe criminal penalties for public servants who misuse public funds or programs to buy votes. Part 5: Election Procedures In response to complaints and proposals from the political parties, IFE has enacted several changes in election day procedures. Through an extended process of trial and error, IFE and the National School for Biological Sciences at the National Polytechnic Institute in Mexico City have developed an indelible ink that will not wash off for at least three days after application. After depositing their ballots in the ballot box, and before leaving the casilla, each voter will have this ink applied to his or her thumb by a bottle with a roller ball. In addition, their voter credentials will be stamped with a small "V" in a marked area corresponding to the 1994 federal elections, indicating that they have already voted in that election. Members of our delegation tested this ink with chlorine bleach, hydrogen peroxide, alcohol, acetone, saliva, Coca-Cola, urine, baking soda, and various household detergents, and found that the government's claim that it was indelible for at least three days was accurate. However, the delegation's test of the voter credential stamp found it to be barely legible, apparently due to the need to protect the laminated card from perforation and oxidation. Members of our delegation tested this ink with chlorine bleach, hydrogen peroxide, alcohol, acetone, saliva, Coca-Cola, baking soda, urine, and various household detergents, and found that the government's claim that it was indelible for at least three days was accurate. Voting booths will be curtained off on all sides to provide for greater secrecy at the time of voting and ballot boxes will be transparent on three sides to prevent them from being stuffed with ballots prior to the start of voting. In addition, every ballot has a corresponding sequentially numbered perforated stub. After voting, and before depositing the ballot in the box, the numbered stub will be detached. This will enable election officials to monitor the exact number of ballots delivered to each precinct and to confirm that the number of reported votes corresponds to the number of validly cast ballots. Polling stations will remain open from 8:00 a.m. until the last person in line before 6:00 p.m. has voted. Voters who are traveling on the day of their election and therefore unable to vote in their own polling stations will have the right to vote for president only in special polling stations. Only a voter credential will be necessary to vote in these special polling stations, which will have no voter registration list. Each district will contain no more than five special stations with the right to receive a maximum of 250 votes each. This translates into roughly 1,500 special polling stations nationwide and a total of 375,000 potential votes at these special stations. Unlike in past elections, election results will be posted outside each polling station at the end of the day, and each political party representative will receive a legible copy. The vote count begins with races for deputy, followed by senator and then president. The count is not open to the public, but party representatives, national observers, and accredited international visitors may be present. Unlike in past elections, election results will be posted outside each polling station at the end of the day, and each political party representative will receive a legible copy. Polling station officials will then deliver an electoral package to the District Council and may be accompanied by party representatives. The electoral package consists of the election results for that station; all used, unused, and annulled ballots in separate envelopes; a document attesting to the opening and closing of the station and any incidents that occurred during the voting process, and any protests by party representatives. Polling stations located within the main town of the district must deliver their electoral packages immediately; those located in smaller towns have 12 hours to do so; and those located in rural areas up to 24 hours. Though district council officials are obliged by law to read aloud the results of each polling station as they arrive, the actual official count will not begin until three days after the polls have opened: the morning of Wednesday, August 24 (COFIPE, Article 246). This is apparently to allow time for the casilla results to be delivered to the district office and also a day for officials to rest.. District officials will begin with the official count of the presidential elections followed by the elections for federal deputy and senator. The results of the presidential elections from the districts will be transmitted to IFE headquarters in Mexico City, where the complete result will be tabulated and issued. Results for senate races will be forwarded to the 32 local councils, which must issue the complete results for their respective states and the Federal District within one week of the elections. The District Councils themselves will declare the winners for the federal deputy races corresponding to the country's 300 single-member districts. The Federal Election Tribunal (TFE), composed of judges proposed by the Mexican President and ratified by a 2/3 vote of the Chamber of Deputies, will rule on any questions regarding the validity of the senate and federal deputy elections. An Electoral College consisting of incoming federal deputies will certify the results of the presidential elections on November 1 when the new Chamber of Deputies is inaugurated. This leaves the General Council of IFE with authority to announce official results only for the Deputies elected by proportional representation (plurinominal) in the five circumscriptions of the country (COFIPE, Article 258). Authoritative unofficial results, however, are expected much sooner. Because the official results will take several days to tabulate, the IFE is planning to do a quick count to generate representative results on the night of the election. They will begin to announce the results of the quick count once a statistically representative sample is available. As actual results arrive they will be included with the quick count announcements. IFE hopes to have authoritative results by midnight on August 21. No one will be allowed to issue preliminary results in the form of a quick count or an exit poll until the polls have officially closed throughout the country. In addition to IFE, Civic Alliance and Televisa, and perhaps three other groups in addition to political parties are planning to conduct a quick count on the day of elections. Views of the Actors The two main opposition parties, the PAN and the PRD, have a number of concerns about election procedures. They are concerned that the provisions of the law might not actually be implemented, particularly in distant polling stations where the opposition may not be able to post a representative. On the question of indelible ink, for example, the opposition parties want to ensure that the ink is not diluted prior to the start of voting. IFE responds that even if the ink is tampered with prior to the start of voting, voters will still be required to have their credentials stamped after voting, thus preventing them from voting twice. All actors favor quick counts in order to reduce uncertainty and discourage fraud from occurring at the counting stage of the process. IFE plans to conduct a quick count in cooperation with the political parties. Close coordination will be necessary to keep secret the identity of the sample sites in order to prevent distortions in the voting pattern from occurring. How IFE plans to balance the challenge of cooperating with the parties to choose the sample while preserving secrecy at the same time is not yet clear. It will also be difficult to choose a representative sample of voting stations given the lack of reliable voting data from past elections. There is the further risk of a war of words over the results of the different quick counts and their respective methodologies that might generate confusion and question the credibility of the electoral process. The PRD and PAN are also concerned that the high number of polling stations and their dispersed location will make it difficult to monitor the electoral process and report results rapidly, particularly in polling stations located in rural areas. They would like to see the country's 97,000 polling stations compressed into 20,000 voting centers. They claim that this could be done in such a way that the vast majority of the citizenry would only be 3 to 4 kilometers away from their polling stations. The matter was proposed to the IFE General Council by Citizen Councilor Josi Woldenberg, and was discussed at length during the Council's June 18, 1994 meeting. No conclusion was reached, and the General Council unanimously requested that the political parties study the feasibility of such a change. The government believes that modifying the location of the polling stations at this late date could generate confusion and discourage voter turnout. If IFE recommends consolidation of polling stations, formal action by the Chamber of Deputies to amend the Electoral Code will be required. Part 6: National Observers The 1993 electoral reforms contained the first law on election observation in Mexico. National observers were to be accredited by IFE as individuals, not as organizations and were required to take a training course given by IFE. The observer must not have been a member of any political party in the three years prior to the election. The law did not guarantee that national observers would be given access to the voter registration list or other information prior to the election. In our last report we noted that this was the most restrictive observer law of which we were aware.In our last report, we noted that this was the most restrictive observer law of which we were aware. The 1994 reforms improve the picture for national observers significantly. The 1994 reforms improve the picture for national observers significantly. Both individuals and organizations can now be accredited by IFE as national observers. IFE will accept training given by the national observer groups as equivalent to its own training course. Observers should not have had a leadership position in a political party at the municipal, state, and national levels in the last three years but may continue to be ordinary members of political parties. In addition, the law guarantees observers the right to whatever information they need for their work as long as it is not confidential by law or beyond the technical and material capacity of the authorities to provide. National observers will be provided with badges to identify themselves on the day of the election. They will be permitted to observe the opening of polling stations, the voting process, the count, the placing of the results outside the station, the closing of the station, and the reading of the results of individual polling stations in the District Councils. The national observers will have the right to submit a report on their work to the General Council but their conclusions will have no legal force. They are explicitly prohibited from declaring victory for any political party or candidate. Views of the Actors The national observer groups seem satisfied with the latest changes to the law on national observers. A variety of observer groups have developed in response to the new law, among them Civic Alliance, an umbrella group of 400 citizen groups; The National Employers Confederation (COPARMEX), a voluntary association of Mexican employers; and the National Union of Teachers (SNTE), a union of public school teachers with ties to the ruling PRI. The Civic Alliance has the most experience in monitoring elections in Mexico but COPARMEX and SNTE have the advantages of a coherent nationwide organizational structure that is already in place and could lend itself to the task of election observation. The crucial challenge facing the national observer groups is to demonstrate their impartiality. The PAN, for example, is worried that the Civic Alliance has pro-PRD leanings, though the PAN does acknowledge that the Civic Alliance has grown more professional over time. The PRI believes that the Civic Alliance is biased in favor of the opposition generally. The Civic Alliance leaders respond that it is a plural organization with members from all three national political parties. They note that there are no party candidates or leaders who hold leadership positions in the Civic Alliance. They also add that it is natural to have a cluster of people with opposition leanings within any civic organization that seeks the democratization of a country. This was the pattern in other democratizing countries like the Philippines and Chile. This, however, does not mean that the organization is incapable of impartiality. The Civic Alliance would like to conduct three kinds of monitoring activity. First, they want to conduct studies of the electoral process itself. The Civic Alliance, as noted earlier, has already conducted a study of bias in the electronic media and plans to continue monitoring the media for the rest of the campaign. The organization would like to conduct its own study of the voter registration list and is also investigating the background of members of the SPE and Citizen Councilors at the district and state levels to identify those individuals who are likely to be biased. In addition, the Civic Alliance is planning to conduct a quick-count on the day of the election and monitor the vote in at least 5,000 polling stations with the aid of 13,000 volunteers. Civic Alliance leaders, however, complained of a severe shortage of funds that is hampering their work, particularly their studies of the electoral process. They also noted that they are battling against the clock with only two months remaining for the elections. Part 7: International Observers The 1993 electoral laws said nothing about the role of international observers. The 1994 reforms, however, empowered the General Council to invite "foreign visitors" to observe the electoral process and establish the guidelines under which foreign visitors were to operate. On June 23, 1994, the General Council issued a set of guidelines regarding foreign visitors.The following groups or individuals are welcome to seek accreditation by the General Council as foreign visitors: representatives of international agencies; regional or continental agencies; legislative organs of other countries; private institutions and nongovernmental organizations defending human rights and/or working on electoral and political issues; specialized groups providing electoral assistance; and foreign personalities known for their contribution to peace and cooperation in human rights and electoral matters. . No observer group may be acting for monetary gain. Political parties and national observer groups may choose to invite various international observer groups if they wish, but IFE will not issue any invitations to any group in particular. Organizations and individuals seeking accreditation as foreign visitors by IFE have until July 30, 1994 to do so and need to complete a prescribed form for that purpose. This is the most bureaucratic process for accreditation of which we are aware. Foreign visitors will have the right to meet with the election authorities and the representatives of political parties in order to obtain information and documents. The General Council will assist foreign visitors to the extent that it is necessary for them to do their task adequately as defined by the law. International visitors will have the right to observe all phases of the electoral process. They will be identified by badges. They are expressly prohibited from interfering in political matters or obstructing the work of the electoral authorities, parties, or candidates, and are exhorted to conduct themselves in a "impartial, serious, and responsible" manner. Violators of these restrictions will have their observer credentials confiscated and will be reported to the Ministry of the Interior for possible deportation. International visitors will have the right to observe all phases of the electoral process. In addition, the government asked the United Nations Electoral Assistance Unit to provide technical assistance to national observer groups. The UN itself commented that it would not consider mounting a formal observer mission given the late decision to invite foreign visitors, the size of the country, and the fact that Mexico does not meet the criteria established by the UN for countries to receive an observer mission. Views of the Actors The PRI and PRD are enthusiastic about the presence of foreign visitors. The PRI hopes that the presence of foreign visitors will provide credibility to the electoral process and results. The PRD believes that the presence of foreign visitors will discourage election fraud and boost voter turnout. The PAN for its part has adopted a welcoming attitude towards foreign visitors experienced in election observation for the same reasons as the PRD. This represents a change from the Council's last visit to Mexico in September, 1993 when the PRI and the PAN seemed ambivalent about the presence of foreign observers and only the PRD was openly enthusiastic. Each of the three main parties have issued invitations to foreign individuals, including some members of the Council and its staff. Part 8: Criminal Penalties; Creation of a Prosecutor's Office for Election Fraud. Unlike the 1993 electoral reforms, the new laws provide for stiff criminal penalties for a gamut of election irregularities. The new laws provide for a prison sentence of six months to three years and a fine equivalent to 10 to 100 times the minimum wage for the following electoral crimes: proselytizing inside the polling station on election day; violating the secrecy of the ballot; interfering with the process of voting; retaining voter credentials without just cause; buying votes in exchange for money or promises of material reward; voting with a false credential; bringing citizens en masse to polling stations with the intention of affecting their vote removing or placing ballots into the ballot box illegally; destroying or altering ballots; obtaining promises of votes through threats, and preventing the opening of polling stations by using violent means. The new laws provide for imprisonment for two to six years and a fine equivalent to 50 to 200 times the minimum wage for the following additional crimes: altering electoral results; removing or destroying ballots and electoral documents; opening or closing a polling station outside the set hours; illegally changing the location of a polling station; expelling political party representatives from the polling stations without just cause, and spreading rumors about the electoral process and results. The new laws provide severe penalties of one to nine years in prison and a fine equivalent to 200 to 400 times the minimum wage for public servants who: use public resources to promote a given party; oblige their employees to vote for a given political party, or make the provision of public services conditional on voting for a party. It is now also a crime subject to three to seven years imprisonment and a fine equivalent to 70 to 200 times the minimum wage to: alter the voter registration list, or illicitly issue a voter credential. The new law also provides for the creation of an Electoral Prosecutor's Office to investigate and punish electoral crimes. The Electoral Prosecutor was to be appointed by the Attorney General after the political parties reached a consensus on the candidate. Only one month before the election, on July 19, the Electoral Prosecutor, Ricardo Franco Guzman, was finally named. Electoral crimes committed before the creation of the Electoral Prosecutor's Office may still be brought before that office for resolution. Views of the Actors The PAN and the PRD applaud both the establishment of criminal penalties for electoral crimes and the creation of an Electoral Prosecutor's Office to prosecute them. In their view, the establishment of criminal penalties deals a blow to the pervasive culture of fraud that has enveloped Mexican elections. The fact that election crimes will now be prosecuted by one single office makes it easier to denounce election crimes and demand follow-up action. The PAN and the PRD, however, lamented the fact that no one had been appointed to the post of Electoral Prosecutor at the time of the delegation visit. Both the PAN and PRD were waiting to submit some complaints until the naming of the Electoral Prosecutor. The government responds that two candidates whom all the parties had already agreed on withdrew for medical reasons, slowing down the process of appointment. PRD presidential candidate Cuauhtimoc Cardenas commented on the new Electoral Prosecutor, "He does not inspire confidence, and I doubt anyone believes he has good intentions." (Unomasuno, July 18, 1994). The PAN and the PRD applaud both the establishment of criminal penalties for electoral crimes and the creation of an Electoral Prosecutor's Office to prosecute them. In their view, the establishment of criminal penalties deals a blow to the pervasive culture of fraud that has enveloped Mexican elections. ANALYSIS AND RECOMMENDATIONS The political and electoral climate has changed markedly since our visit in September 1993. New electoral reforms stimulated by civic pressure, the rebellion in Chiapas, and the Colosio assassination have led to a growing independence of electoral authorities and improvements in electoral procedures. Yet the absence of a clear front-runner in the presidential race, the climate of uneasiness in the wake of a high-profile assassination and the Chiapas rebellion, and the continued skepticism of the Mexican voters have produced a fluid and potentially volatile atmosphere surrounding the upcoming elections. The two major opposition parties indicated that while the rules were now in place for a free and fair election, they questioned whether the political will exists to implement and enforce them. They questioned the ability of the new IFE General Council to exert any real control over a hierarchy of thousands of election officials in the time remaining before the election. They pointed to continued bias in media coverage that favors the PRI, the cost of advertising that makes it difficult for them to compete, and vast disparities in financial resources among the parties. The PRD in particular remained concerned about the reliability of the voter registration list even after the results of the audits were known . Therefore, neither the PRD nor the PAN were prepared in June to say definitively that they could accept the election results. In our 1993 report, we separated the electoral issues into first-order issues necessary to have a meaningful election and second-order issues desirable to have a good election. First-order issues include (1) the political parties must have a chance to communicate their message to the people free of intimidation; (2) eligible voters must have an opportunity to register and cast their vote in a secret ballot (implying a good registration list, but especially one that is not politically biased); (3) the count must be fair and accurate; (4) opposition poll-watchers and independent monitors should have complete access to every stage of the vote and count; and (5) effective power must be transferred to the winner. Second-order electoral issues include (1) a fair, but not necessarily equal, distribution of campaign resources; (2) equitable access to the media; (3) an agreed-upon formula for distributing state resources to the parties in a transparent way; and (4) an independent and impartial election administration. First-Order Issues Our analysis of the electoral process as of June 1994 finds that there has been substantial progress on some of the first-order issues, particularly with regard to the technical aspects of the election. The crucial issues in past Mexican elections have related to the voter registration list and voting process and count. The Federal Electoral Registry (RFE) has mounted an impressive and costly effort to purify the voter's list and deliver new tamper-proof credentials to 90% of the eligible voters. The coverage of the voter's list (the percentage of eligible voters actually registered) is thus comparable to most Western democracies. The reliability of the list also depends on two other criteria, however: currency and accuracy. Our preliminary conclusion is that the Padrsn provides a good basis for a free election. Nevertheless, we find the conclusion that the Padrsn is 97.5% reliable to be the most optimistic interpretation of the data. It is also confusing because it mixes the two criteria of currency and accuracy. The opposition parties have identified two possible avenues for fraud related to the list. The McKinsey audit addresses one of those issues: the potential for padding the list with fantasmas or nonexistent persons is small. However, the audit does not directly address the second issue of the possible disfranchisement or exclusion of voters who believe themselves to be registered and qualified. Consequently, other safeguards must be relied on to help protect against these concerns: vigilant use of indelible ink to prevent multiple voting, and the review of the final voter's list to ensure that all registered voters are in fact on the list and permitted to vote. The vote and vote count processes have been improved by the changes in the election day procedures, including the ballot boxes, voting booth, and ballot paper, as well as by the significant improvement in the regulations of national observers and the decision to accredit foreign visitors to observe the process. In addition, the designation of a special prosecutor for electoral crimes and the new criminal code are important steps toward ensuring a free and fair election. The late decision to invite foreigners, however, makes impossible a traditional observer mission capable of monitoring the entire process. Likewise, the delay in naming the special prosecutor impedes the effectiveness of this reform. Second-Order Issues Second-order issues have assumed greater importance in Mexico than in many other transitional countries because of the tremendous power of the ruling party and the difficulty of separating it from the state. These issues determine how level the playing field is and remain of real concern in the Mexican process. Continued problems of media access and bias in coverage, campaign spending limits beyond the reach of any party outside of the PRI, and great disparities in financial resources all affect the ability of parties to compete fairly. The greater independence of IFE's General Council is a marked improvement; however, the ability of the General Council at the apex to influence the actions of the rest of the pyramid of the electoral machinery is severely limited. In most countries, the actual independence of the electoral authorities from government control is less important than impartiality in decision-making and behavior. In Mexico, the long-term blurring of party and state, and the lack of third-party mediators raises the independence of IFE almost to a first-order issue. The fact that the political parties put their faith in Council President and Interior Minister Jorge Carpizo is understandable given his personal reputation, but it also reflects the continued perception that the government does in fact control the process. In every country, the incumbent party has an advantage deriving from it's control over state programs. On the other hand, the incumbent party also has a liability in that voters can judge actual performance and not just future promises. In the Mexican case, however, the question of a level playing field is particularly acute given the total hegemony of a single party over government resources and power for such a long period of time. Not only does the PRI control government decisions and resources, but its penetration of civil society and close ties to the private electronic media give it important additional sources of influence. An uneven playing field can be detrimental rather than advantageous to the ruling party in some cases. But in Mexico it reinforces the image of the PRI as controlling the process, and discourages voters from thinking that the election will matter or their votes will count. This is expressed in the fears of intimidation of voters by employers, trade union leaders, or government officials; fears of intimidation on election day, such as the visible presence of the military in the streets, that will curtail voter turnout; fears of locally-based fraud and intimidation in isolated areas without observers; and fears of post-election violence. These fears all reflect the skepticism of the Mexican populace toward the ruling party's commitment to competitive elections. They reflect a prevalent view expressed to the delegation that certain elements of the ruling party and privileged elite will go to almost any length to prevent a defeat at the polls, and therefore a loss of power and privilege. As one government official commented to the delegation, people can say they believe in the electoral rules and procedures, but then they turn around and say they expect fraud. These attitudes raise a troubling question: beyond the issue of a free and fair election, how meaningful is an election in which a plurality of voters doubts that their votes will count? Will those voters who lack confidence stay away from the polls or vote other than their true choice because they believe their vote will not be respected? If that is the case, even the most technically beautiful election will not reflect the will of the people. These attitudes raise a troubling question: beyond the issue of a free and fair election, how meaningful is an election in which a plurality of voters doubts that their votes will count? Will those voters who lack confidence stay away from the polls or vote other than their true choice because they believe their vote will not be respected? We believe there has been substantial progress in creating the conditions for a free and fair election. Nonetheless, the continued distrust of nearly half of the voters raises questions about whether the elections will be meaningful. The wait-and-see attitude of the PAN and the highly suspicious attitude of the PRD further raises the question of whether the results will be accepted by all the major parties. We recommend a set of actions that can help deal with the problem of credibility. Recommendations: I. Improving Credibility 1. To build credibility, additional debates should be held involving the three major presidential candidates. Of all the factors affecting public confidence, the presidential candidates' debate appears to have had the most positive effect. 2. The leaders of the three major political parties should reach out to each other and toward the people to help build confidence in the process by: (a) jointly filming public service announcements on the need to vote and how to vote; (b) discussing ways to ensure that the results will be announced soon after the election and that the transition would be as smooth as possible. 3. IFE should conduct an extensive civic education campaign to explain the procedures and safeguards, and identify the location of each voting site, which in turn should help raise confidence in the process. II. First-Order Issues 4. The final voters list should be posted by casilla in the ten days preceding the election to allow voters to confirm both their inclusion on the list and the location of their casilla. This should also help political parties and observers determine if there is any systematic exclusion of voters from the list. 5. Ensure a rapid and efficient transmission and announcement of unofficial results on election night. This will help greatly to alleviate suspicion and maintain calm. 6. Quick counts, or parallel vote tabulations, are important instruments to verify official results and to provide early information about voting trends. We believe that exit polls, on the other hand, should be avoided. They are unreliable in a climate of suspicion and they will create a negative atmosphere if voters think they are being watched. 7. We commend the political parties' efforts to organize party representatives to be present in every casilla (voting site). To the extent that this is not feasible for each party, we recommend a coordinated strategy to ensure that there are at least two parties represented in each casilla. 8. National observers should undertake a mobile strategy on election day to cover as many casillas as possible, making a special effort to have representation at every polling site where there are no opposition pollwatchers. 9. International visitors should be publicly welcomed by political party leaders to overcome an ambivalent and sometimes antagonistic climate. The UN should coordinate international visitors to the extent possible to ensure balanced coverage of the entire country. 10. To offset lingering concerns about multiple voting, the ink should be guarded until election day and samples tested publicly immediately before distribution to ensure that it has not been tampered with. A cleanser may be desirable to clean and dry fingers before they are inked. Party representatives and observers should be especially vigilant to determine if voters' fingers are properly inspected and inked. III. Second-Order Issues 11. The role of state and district officials is vital in determining the electoral results. In the future, additional district- and state-level Citizen Councilors should be replaced. 12. We commend IFE's analysis and publication of reports on media coverage, but bias remains. The decision of Televisa to donate 45 minutes to each of the nine presidential candidates for political broadcasts should be matched by TV Azteca and by radio. Some of that donated time should be during primetime. 13. Political parties should make voluntary pubic reports of revenues and campaign expenditures before the election even though the law requires reporting only after the election.CONCLUSION In our 1993 report, we concluded that Mexico has two roads available - one leading to a new formula for dividing power, the other leading toward full respect for the secret vote and acceptance of the uncertain outcome that is a part of the democratic process. A negotiated outcome would lead down the first road and could delay the search for democracy. The rules have been put in place for the second road -- a secure vote -- but continued distrust of the process combined with an increased focus on post-election negotiation detracts from a free election. Those interested in post-election scenarios may want to contribute to a political opening, but in fact, their perspective of negotiated power-sharing arrangements or street protests represent a new variation on an old theme of depriving the people of a chance to choose their government in a secret vote. Unless confidence in the electoral process is raised before August 21, the vote may be technically clean, but will not necessarily represent a meaningful election. Because we have followed closely the electoral process for several years and because the Council aims to promote democracy, peace, and hemispheric cooperation, we want to be present for the August 21 elections. We will not field an election team like those we have sent to other countries. Instead, we will send a small team that will coordinate with others and will reinforce the efforts of Mexican observer groups and all those working to make these the freest elections in Mexican history, not tainted by manipulation or fraud, and to reassure Mexican voters that the international community supports their right to exercise their free choice through the ballot box. APPENDICES Appendix 1: List of Meetings Mexican Civic Monitoring Groups Sergio Aguayo, Mexican Academy for Human Rights/ Civic Alliance Gina Batista, ACUDE Francisco Cano Escalante, ACUDE, Council for Democracy Julio Faesler, Council for Democracy/ Civic Alliance Clara Jusidman, ACUDE Luz Rosales, Citizens Movement for Democracy Citizen Councilors Santiago Creel Josi Agustmn Ortmz Pinchetti Ricardo Pasos Josi Woldenberg Party of Democratic Revolution (PRD) Amalia Garcia, Secretary of International Relations Javier Gonzalez, Secretary of Electoral Affairs Porfirio Muqoz Ledo, President of the PRD/ Senator Ricardo Pascoe, Advisor to Cuauhtimoc Cardenas Jeszs Zambrano, PRD Representative to the Federal Election Registry National Action Party (PAN) Felipe Caldersn Hinojosa, Secretary General of the PAN Rodolfo Elizondo Torres, Secretary of Relations Cecilia Romero Castillo, Adjunct Secretary General Antonio Losano, Representative to IFE Josi Lums Luege Tamargo, Director of RFE Relations Institutional Revolutionary Party (PRI) Rodolfo Becerril Straffon, Chamber of Deputies; Senate Candidate Enrique Berruga, Undersecretary for International Affairs Emilio Cardenas Mauricio Reyes Federal Election Institute (IFE) Carlos Almada, Executive Director, Federal Election Registry Manuel Carrillo Poblano, Coordinator, International Relations Juan Molinar Horcasitas, Director of Political Parties and Prerogatives Arturo Nuqez Jimenez, Director General of IFE Agustmn Ricoy Saldaqa, General Secretary, Federal Election Registry Felipe Solms Acero, Director of Electoral Organization Esteban Garcma I., Director, Regional Coordination The Presidency Ulises Beltran, Advisor to President Carlos Salinas de Gortari Beatriz Paredes R., Undersecretary for Political Development, Ministry of Gobernacisn Susana Torres, Director for Political Development Emilio Robase, Advisor Independent Analysts Jorge Castaqeda G. Miguel Basaqez, Este Pams/MORI Federico Reyes Heroles, Este Pams United Nations Dong Huu Nguyen, Director, UN Office for Electoral Assistance - Mexico Francesco Manca, Senior Technical Advisor Araceli Santana, Senior Technical Advisor COPARMEX Marco Antonio Velazquez, Director General Lorenzo PalaisAppendix 2: Biographies of Delegation Members Dr. Jennifer McCoy, Senior Research Associate of the Carter Center of Emory University's Latin American and Caribbean Program. Professor of Political Science at Georgia State University. Led the Council's September 1993 and June 1994 missions to assess Mexico's electoral reforms, and directed Council's Nicaragua election office and has observed election in Haiti, Guyana, Panama, and Suriname. Hon. Peter Stollery, Representative of Council member former Canadian prime minister Pierre Trudeau. Elected House of Commons 1972, 1974, 1979, 1980; Chairman, Government Caucus 1976-78; Appointed to Senate of Canada by Prime Minister Trudeau, 1981; Founder and First Chairman, Canada-Latin America Parliamentary Group. Mr. Joaqumn Daly, representative of former Peruvian president Fernando Belaunde Terry. Former official of the OAS, and electoral consultant to The Carter Center. Participated in the Council mission to witness Mexican observers in the July 1992 state elections in Michoacan and Chihuahua, and in the Council's election observation mission in Panama in 1994. Dr. Vikram Chand, Watson Fellow in International Relations at Brown University; specialist in Mexican politics and democratization. Participated in Council mission to assess electoral reform in Mexico in September 1993. Mr. Harry Neufeld, Canadian management consultant in electoral systems and information technology. Former Director of Information Technology in Elections Canada. Participated in Council pre-election assessment missions to Guyana in 1991. Mr. Eric Bord, attorney and consultant to The Carter Center. Former Assistant Director of the Latin American and Caribbean Program of The Carter Center, and consultant to Carter Center programs in Conflict Resolution and Global Development Cooperation. Participated in Council election observations in Nicaragua and Guyana. 1 There is substantial controversy in Mexico over polling methodology and sample size. We refer to these polls for the purpose of illustrating and underscoring the problem of credibility in the 1994 elections, and with no intention to endorse any particular methodology. 2 In Mexico, the Padrsn refers to the list of registered voters while the Listado Nominal Definitiva refers to the list of those voters who both registered and received a credential. It is the Listado Nominal Definitiva that is used on election day. 3 The PDM later clarified that its abstention was "provisional" pending the conclusion of its own independent study of the audit results.