Observing the 2004 Mozambique Elections - The Carter · PDF fileObserving the 2004 Mozambique Elections Waging Peace. Fighting Disease. ... São Sebastião Cahora Bassa Dam Licoma

SPECIAL REPORT SERIES

Observing the 2004Mozambique Elections

Waging Peace. Fighting Disease. Building Hope.

THE CARTER CENTER STRIVES TO RELIEVE SUFFERING

BY ADVANCING PEACE AND HEALTH WORLDWIDE;IT SEEKS TO PREVENT AND RESOLVE CONFLICTS, ENHANCE FREEDOM AND DEMOCRACY,

AND PROTECT AND PROMOTE HUMAN RIGHTS WORLDWIDE.

OBSERVING THE 2004MOZAMBIQUE ELECTIONS

ONE COPENHILL

453 FREEDOM PARKWAY

ATLANTA, GA 30307

404-420-5188FAX 404-420-5196

WWW.CARTERCENTER.ORG

OCTOBER 2005

THE CARTER CENTER

OBSERVING THE 2004 MOZAMBIQUE ELECTIONS

2

Delegation and Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Terms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14

The Carter Center in Mozambique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

The 1999 Elections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

The 2003 Municipal Elections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Electoral Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Voter Registration Update . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26

Campaign Finance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30

Pre-election Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36

Election Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44

Postelection Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51

Partnership with Civil Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56

Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63

Appendices

a. CNE Invitation to Observe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .67

b. Short-term Deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68

c. Observation Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70

d. Carter Center Public Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76

The Carter Center at a Glance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90

TABLE OF CONTENTS

THE CARTER CENTER


3

MorrumbalaNamacurra

Marromeu

Vila Fontes

Inhaminga

MopeiaVelha

Chinde

Vila da Maganja

AngocheGile

Nacala

Mogincual

MossurilRibáuè

Moçambique

NametilMoatize

Milange

MurrupulaMeconta

Namarrói

MecubúriCuamba

Erego

Lugela

Memba

Quissanga

Chiconono

Maúa

Namuno

Mecula

Ancuabe

MontepuezMarrupa

Metangula

Namapa

Lichinga

Mocimboa da PraiaMueda

Palma

Vila Coutinho

Changara

Magué

Zumbo

Manica

Sussundenga

FurancungoFíngoè

Vila Gamito

Guro

Mandimba

Chemba

Chibabava

Dondo

Vilanculos

Nova Mambone

Morrumbene

Massinga

Mabote

Mapai

Espungabera

Chicualacuala

PandaHomoine

Quissico

Inharrime

Magude

MoambaManhiça

ChibitoGuija

Massingir

Namaacha

Bela Vista

Inhassôro

Pebane

Moma

VilaJunqueiro

Macia

Manjacaze

Monapo

Boane

Pemba

Nampula

Quelimane

Chimoio

Tete

Beira

Inhambane

Xai-Xai

Lilongwe

Harare

Maputo

Mbabane

Pretoria

Lusaka

ZAMBÉZIA

MANICA

GAZA

MAPUTO

INHAMBANE

CABODELGADO

N I A S S A

T E T E

S O F A L A

N A M P U L A

Ilha do BazarutoIlha Benguérua

Ponta da Barra Falsa

Ilha Angoche

Cabo Delgado

Ponta São Sebastião

Cahora BassaDam

Licoma Is.

ZIMBABWE

BOTSWANA

Z A M B I A

SOUTH AFRICA

SWAZILAND

DEMOCRATICREPUBLIC OFTHE CONGO

UNITED REPUBLIC OF

TANZANIA

MA

LA

WI

(MALAWI)

INDIAN

OCEAN

I N D I A N O C E A N

Mo

za

mb

iq

ue

Ch

a n n e l

Lake Kariba

LakeMalombe

Lago deCahora Bassa

LakeBangweulu

LakeChifungwe

LakeChilwa

Limpopo

Changane

Save

Buzi

ZambezeLicungo

Ligonha

Lúrio

Rovuma

Lugenda

Ruvuma

Messalo

La

ke Ma

law

iL

ake Nyasa

Luan

gwa

Limpo

po

Zambezi

Olifants

50 150100 200 km0

0 50 100 150 mi

Map No. 3706 Rev. 5 UNITED NATIONSJune 2004

Department of Peacekeeping OperationsCartographic Section

The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations.

MOZAMBIQUE

National capital

Provincial capital

Town, village

Airport

International boundary

Provincial boundary

Main road

Railroad

MOZAMBIQUE

THE CARTER CENTER


4

DELEGATION AND STAFF

DELEGATION LEADERS

The Honorable Jimmy Carter, 39th President of the United States, Carter Center ChairMrs. Rosalynn Carter, Former First Lady of the United States, Carter Center Vice ChairThe Honorable Nicéphore Soglo, Former President of BeninDr. John Hardman, Executive Director, The Carter Center, United States

DELEGATION

Helena Alves, Election Consultant, PortugalBradley Austin, IFES, United StatesPatrick Berg, Graduate Student, GermanyJason Calder, Assistant Director, The Carter Center, United StatesBecky Carter, Director of Government Relations, The Nature Conservancy, United StatesChip Carter, Vice President, Primacy International, United StatesWellington Chibebe, Secretary-general, Zimbabwe Congress of Trade Unions, ZimbabweGlenn Cowan, Consultant, Democracy International, United States Mario de Paiva, Journalist, AngolaMarc de Tollenaere, Senior Program Officer, European Center for Development Policy Management, Former Carter Center Representative in Mozambique, BelgiumAmanda Dixon, Graduate Student, Georgetown University, United StatesElma Doeleman, Netherlands Institute for Southern Africa, The NetherlandsJohn Fleming, Freelance Writer, United StatesTorben Vestergaard Frandsen, Director of Development, Vestergaard Frandsen, DenmarkNina Frankel, International Health and Social Sector Specialist, United StatesAna Ganho, Assistant Professor, Emory University, PortugalCipriano Gomes, Doctoral Candidate, University of Stockholm, Guinea BissauMargot Gould, Assistant Program Officer, International Institute for Democracy and Electoral Assistance, AustraliaKathleen Hawthorne, Former Director of Engineering, Bell South, United StatesFrances Henderson, Doctoral Candidate, Columbia University, United StatesSamuel Kivuitu, Chair, Electoral Commission of Kenya, KenyaBill Kleh, Attorney, Trustee, The Carter Centre United Kingdom, United StatesMaria Macchiaverna, Elections Specialist, ItalyCarrie Manning, Professor, Department of Political Science, Georgia State University, United StatesRene Mongbé, Member of National Assembly, BeninFrances Johnson Morris, Chair, National Electoral Commission, LiberiaJean Paul Murekezi, Researcher, RwandaJane Nandy, Senior Associate Director of Development, The Carter Centre, United KingdomGemima Neves, Program Officer, National Democratic Institute, United StatesAchille Nisengwe, Researcher, Rwanda

THE CARTER CENTER


5

Paul Nsapu, National President, League of Voters, Democratic Republic of the CongoAnne Pitcher, Professor, Colgate University, United StatesMirita Rodrigues, Education Program Specialist, UNESCO, AngolaScott Taylor, Professor, African Studies Program, Georgetown University, United StatesGirum Tesfaye, Student, Winnipeg University, EthiopiaHenry Tinsley, Trustee, The Carter Centre United Kingdom, United KingdomRebecca Tinsley, Trustee, The Carter Centre United Kingdom, United KingdomHarry Vanden, Department of Government and International Affairs, University of South Florida, United States Mooroogessen Veerasamy, Principal Election Officer, National Elections Commission, Mauritius

The Carter Center delegation to Mozambique was comprised of 60 members from 23 countries.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


6

LONG-TERM OBSERVERS

Anna Bjorndal, NorwayRoel Borren, The NetherlandsNatasha Cassinath, CanadaMário Jaleco, PortugalAbdoulaye Korouma, GuineaFernanda Lópes, PortugalCecília A. Luna López, EcuadorJacques Saidi-Kamuleta, Democratic Republic of the CongoSilvina Silva-Aras, Argentina

THE CARTER CENTER STAFF –ATLANTATom Eberhart, Associate Director of Finance, The Carter Center, United StatesBob Ellzey, Leadership Coordinator, The Carter Center, United StatesSimona Foltyn, Intern, The Carter Center, AustriaTynesha Green, Program Assistant, The Carter Center, United StatesNancy Konigsmark, Assistant to President Carter, United StatesDan Kosinski, Intern, The Carter Center, United StatesNealin Parker, Assistant Program Coordinator, The Carter Center, United StatesDavid Pottie, Senior Program Associate, The Carter Center, CanadaKay Torrance, Assistant Director of Public Information, The Carter Center, United States

MAPUTO STAFFNicolás F. Bravo, Field Office Director, The Carter Center, ArgentinaJúlio Garcia Cruz, Logistics Coordinator, The Carter Center, United StatesFátima Mahel, Finance Assistant, The Carter Center, MozambiqueAlda Mahumane, Secretary, The Carter Center, MozambiqueDionisio Spiratus, Driver, The Carter Center, MozambiqueDaude Sulemane, Driver, The Carter Center, Mozambique

THE CARTER CENTER


7

Acta Formal minutes of the presiding offi-cers of a polling station

AMODE Associação Moçambicana Para oDesenvolvimento da Democracia(Mozambique Association for theDevelopment of Democracy), domes-tic observation organization

Cadernos Segments (or books) of the voter reg-ister with the names of registeredvoters that correspond to a pollingstation

CNE Comissão Nacional de Eleições(National Elections Commission)Responsible for supervision of elec-tion administration and composed ofmembers from Frelimo and Renamo

Commonwealth The British Commonwealth is anassociation of 53 countries that spon-sored an international observationdelegation for the 2004 elections

CPE Comissão Provincial de Eleições(Provincial Elections Commission),provincial branch of the CNE

Delgados Delgados de Candaturas, partypolling station agents

DfID Department for InternationalDevelopment, based in the UnitedKingdom

Editais Official tally sheets handwritten atthe end of the voting process at eachvoting table (singular: edital)

EISA Electoral Institute of Southern Africa

EO Electoral Observatory

EU European Union

FECIV Forum de Educação Cívica (CivicEducation Forum)

Fiscais Monitors: fiscais dos partidos refersto political party members monitor-ing the election

Frelimo Frente da Libertação de Moçambique(Liberation Front of Mozambique)

Mesa A voting table consisting of five pollworkers administering the poll

OE Observatório Eleitoral (ElectoralObservatory, EO)

PVT Parallel vote tabulation (Recolha deApuramentos Parciais -RAP)

Renamo-UE Resistência Nacional Moçambicana–União Eleitoral (MozambiqueNational Resistance –ElectoralUnion)

SADC PF Southern African DevelopmentCommunity Parliamentary Forum

STAE Secretariado Técnico deAdministração Eleitoral (TechnicalSecretariat for ElectoralAdministration) The administrativearm responsible for conduct of elec-tions, including training of pollingofficials and civic education

USAID United States Agency forInternational Development

TERMS AND ABBREVIATIONS

THE CARTER CENTER


8

Over the course of 2003-2004, the Centerworked with election authorities, domesticelection observers, and members of the inter-

national community to strengthen Mozambique’selectoral process. I was especially pleased to see our col-laboration with domestic observers produce thecountry’s first parallel vote tabulation of presidentialand legislative election results. This independent checkon the official results helped to build confidence inthe final outcome.

The Carter Center organized a comprehensiveobservation program of the electoral process, includingan assessment of voter registration and the deploymentof long-term observers to monitor the campaign peri-od. For the Dec. 1-2 elections, the Center deployed a60-person delegation, and long-term observersremained for extended monitoring of the vote tabula-tion process. Taken as a whole, the 2004 electoral

process demonstrated a number of positive signs,including a generally peaceful campaign period andvoting process and improved accountability during thevote counting.

Unfortunately, as in 1999, technical problems anda lack of transparency in the final tabulation of nation-al results delayed the announcement of results andundermined the credibility of the process. These prob-lems prevented the Center from concluding with anentirely positive assessment of the election process.The National Elections Commission and the electionadministration staff must redouble their efforts toensure that future elections in Mozambique are con-ducted on the basis of a credible voter register, secureand transparent administrative processes, and timelyannouncement of final results.

President Armando Guebuza and his party,Frelimo, are to be congratulated on their convincingvictory at the polls. Their challenge will be to work withthe main opposition party Renamo, other parties, andcivil society actors to find political solutions to theproblems that the Center and others have observed.

I want to extend special thanks to NicéphoreSoglo, former president of Benin, for joining me as aco-leader of the December 2004 delegation. His experi-ence and wisdom added immeasurably to our efforts.The Carter Center is especially grateful to the UnitedKingdom’s Department for International Developmentand the United States Agency for InternationalDevelopment, whose generous support made this ini-tiative possible. We also appreciate the fundingprovided by Switzerland’s Federal Department ofForeign Affairs.

FOREWORD

Jimmy Carter addresses the media following a meeting withoutgoing President of Mozambique Joaquim Chissano.Rosalynn Carter is in the immediate background.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


9

The Carter Center would like to thank a num-ber of individuals and organizations withoutwhose tremendous support and efforts the

Mozambique electoral process observations over thelast five years would have been impossible.

The Center thanks the government ofMozambique and the National Elections Commissionpresident, Reverend Arão Litsure, for inviting theCenter to observe the elections. The Center acknowl-edges that the 2004 elections were a substantialundertaking, requiring the coordination of the CNE,STAE, government officials, and the internationalcommunity.

The Center acknowledges the support of theUnited States Agency for International Development,the United Kingdom Department for InternationalDevelopment, and Switzerland’s Federal Departmentof Foreign Affairs. Their combined support enabledthe Center to implement a sustained program of activ-ities from the 2003 municipal elections through thevoter registration update to the 2004 presidential andlegislative elections. The Center would like to extendspecial thanks to Miguel de Brito of USAID, AliciaHerbert of DfID, and Anne Gloor of the Embassy ofSwitzerland for their assistance.

The Center offers special gratitude to formerPresident of Benin Nicéphore Soglo, who served asdelegation co-leader with President and Mrs. Carter tothe presidential and legislative election observations.President Soglo has responded to the Center’s invita-tions on many previous occasions, and his insightsand experience strengthened our work inMozambique.

The Carter Center was fortunate to have two dedi-cated and talented Mozambique office directors. Marcde Tollenaere directed the office through the municipalelections and voter registration update and continuedto participate in the project after his departure from

Mozambique in August 2004. Mr. de Tollenaere was acentral figure not only in this project but also in thewider political scene in Mozambique, and it is a sign ofrespect among Mozambicans for his intelligence anddiligence that he was able to collaborate so effectivelywith the Center’s domestic partners. Nicolás Bravostepped into his role as the new field office director inSeptember 2004 and immediately demonstrated that hewas up to the many challenges ahead. Mr. Bravo organ-ized the Center’s long-term observer deployment,continued the collaboration with domestic electionobservers and support to the PVT, managed thearrangements for a 60-person election delegation andtwo former presidents, and helped the Center to remainengaged throughout the lengthy postelection process.The Carter Center is indebted to the leadership andinsight of these two individuals.

Our Maputo office staff was remarkable not onlyfor their competence but also their unwaveringpatience. Cecília Luna López coordinated the long-term observation and planned for deployment of theshort-term observation; Júlio Garcia Cruz coordinatedlogistics for the delegation leadership; Fátima Maheldoubled her effort as finance officer and deploymentarranger; Alda Mahumane provided continuous sup-port in the office; and Dionisio Spiratus and DaudeSulemane served professionally as drivers with an inti-mate knowledge of Maputo.

Long-term observers bear the brunt of observingand compiling election information for weeks beforeand after the delegation’s arrival and departure.Cecília Luna López, Amanda Dixon, AbdoulayeKourouma, and Jacques Saidi-Kamuleta workedthroughout the voter registration update. They werejoined by Anna Bjorndal, Roel Borren, NatashaCassinath, Mário Jaleco, Fernanda Lópes, and SilvinaSilva-Aras prior to the December elections. Each ofthem traveled to multiple locations in Mozambique,

ACKNOWLEDGMENTS

THE CARTER CENTER


10

covering all provinces, and without their hard workand insights, the Center would not have been able togain the same understanding of the elections.

The Carter Center thanks the individuals who vol-unteered their time and experience to join thedelegation in Mozambique. They accepted theirresponsibilities without complaint and demonstratedtheir support for democracy in Mozambique. The dele-gation had the pleasure of counting several CarterCentre United Kingdom trustees as observers: Henryand Rebecca Tinsley and Bill Kleh. The Center wouldlike to offer special thanks to Torben VestergaardFrandsen for his participation and contribution ofmedicated mosquito nets for the entire delegation.

Several consultants provided special skills to theproject: Glenn Cowan and Eric Bjornlund (parallelvote tabulation) and Bruno Speke (campaign finance).Their expertise and ability to appreciate the specificconditions of Mozambique were crucial to the successof these activities.

A number of Carter Center staff worked fromAtlanta to make the observation possible, includingTynesha Green, Tom Eberhart, Kay Torrance, NancyKonigsmark, Jane Nandy, Jason Calder, and Shelley

McConnell. Several interns contributed in countlessways: Dan Kosinski, Simona Foltyn, RicardoRodrigues, Amy Cook, and Amanda Dixon. They con-tributed to logistics, research, and preparation ofmaterials for the delegation and traveled toMozambique during the elections. Without this hand-ful of staff and interns alike, the Center would havebeen unable to implement this project.

The Carter Center recognizes the work of all theMozambican national observers, especially BrazãoMazula and Otilia Aquino for their leadership in theElectoral Observatory. Domingos do Rosario is com-mended for his stewardship of an especiallydemanding and crucial role in the PVT central office.Likewise the Center acknowledges the work of otherinternational observers, particularly the delegationsfrom the European Union, Commonwealth, andElectoral Institute of Southern Africa.

Many individuals contributed to this report–Nicolás Bravo, Glenn Cowan, Dan Kosinski, NealinParker, and Marc de Tollenaere–based on input fromthe entire delegation. David Pottie managed the overallproject and compiled the final version of the report.

Members of the Carter Center core staff in Mozambique with President and Mrs. Carter.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


11

1.Following nearly two decades of civil war,Mozambique held its first democratic elec-tions in 1994. Widely determined to be free,

fair, and successful, the election illustrated the extentto which two main forces, Frelimo and Renamo, hadcommitted themselves to peace and the introductionof multiparty politics. The Carter Center observed the1999 elections in Mozambique. These elections werelargely peaceful, and election day procedures were well-administered. The Center had serious concerns abouta lack of transparency and observer access during thetabulation process. The narrow margin of victory forPresident Chissano over challenger Afonso Dhlakamacoupled with a significant number of rejected pollingstation tallies cast a lingering shadow over the elec-tions. Postelection conflict between the two partiesthreatened to undermine the hard-won peace inMozambique.

2.The National Elections Commission invitedThe Carter Center to observe the November2003 municipal elections, and the Center

opened an office in Maputo in October 2003. Areport of the Center’s observations is available on theCenter’s Web site. The Center collaborated withdomestic election observers to conduct a parallel votetabulation that serves as an independent check on theofficial results.

3.The Carter Center remained in Mozambiquethroughout 2004, working with the domesticobservers and observing the voter registration

update conducted in June-July 2004. The Centerraised several important concerns regarding the credi-bility of the voter register, which held many duplicateentries and names of the dead. A report is available onthe Center’s Web site.

4.The CNE extended an invitation to PresidentCarter and The Carter Center to observe theDecember 2004 elections. Prior to the elec-

tions, the Center deployed long-term observers whovisited more than 50 districts and every province. Theymet with representatives of the political parties, elec-toral authorities, domestic and international electionobservers, civic organizations, media, and internationalcommunity representatives. Generally, observers founda calm environment, although some isolated signs ofintimidation (some involving the police force) wereobserved in Gaza and Tete provinces. The Centerhoped that the Technical Secretariat for ElectoralAdministration would intensify its voter educationcampaign in the remaining weeks before the elections.The Center observed the voter registration update inJune-July and expressed serious concerns regarding theaccuracy of the voter register; these concerns persistedduring the campaign period. No final list identifyingthe number of registered voters per polling station wasmade available to political parties or observers.

5.The Center deployed a delegation of 60observers from 23 countries for the electionson Dec. 1-2. Following briefings in Maputo,

observers were deployed in teams to every province. TheCenter coordinated its efforts with other internationaland domestic observer organizations. The delegates metwith local and provincial election officials. PresidentJimmy Carter, Mrs. Rosalynn Carter, and former BeninPresident Nicéphore Soglo led the delegation. The exec-utive director of The Carter Center, Dr. John Hardman;senior program associate and Mozambique project man-ager Dr. David Pottie; and the Center’s Mozambiquefield representative, Mr. Nicolás Bravo, accompaniedthe delegation leaders. The leadership team remained inMaputo, meeting with President Joaquim Chissano; thepresidential candidates; members of the CNE, STAE,and Constitutional Council; leaders of nonpartisan

EXECUTIVE SUMMARY

THE CARTER CENTER


12

domestic election monitoring organizations; leaders ofother international election observation delegations;and others.

6.Over the course of polling on Dec. 1-2, theCenter’s observers found a calm environment,careful attention to procedure by polling offi-

cials, and a notable number of women polling officials.In general, the Center’s observers felt they were wel-comed and able to observe freely. Despite the Center’sconcerns about the overall quality of the voter register,eligible voters appeared to be able to cast their ballotswithout difficulty. Voter turnout appeared to be low(less than 40 percent), and several serious irregularitieswere observed during the counting process in Niassaand Tete. The newly introduced provision for partyand candidate agents to sign and receive copies ofpolling station tally sheets was a welcome step toward amore accountable results process.

7.Evidence of serious irregularities in the pollingprocess came to light in several provinces. Forexample, the Center’s observers in the Tete

provincial districts of Changara, Chifunde, andTsangano as well as in the Niassa districts of Metaricaand Marrupa and in the Gaza district of Chicualacualafound voter turnout percentages suspiciously high andin some cases, impossibly high (more than 100 per-cent), leading to the conclusion that ballot stuffingoccurred in some of those polling stations. These inci-dents had an important impact in the final NationalAssembly results. Also notable, the province of Tetehad the highest (and unprecedented) voter turnoutnationwide (67.4 percent), contrasting with a nationalrate of 43.6 percent.

8.The official results process was delayed. Noprovince met the legal deadline for announce-ment of interim results, and the CNE

declared final results on Dec. 21, four days after the

deadline. The Center acknowledges that the multipartyenvironment in Mozambique provides an importantopportunity for parties to work together to improvegovernance, but the public also has a right to theprompt and transparent verification and announce-ment of results. Approximately 3.3 millionMozambicans, or slightly more than 40 percent of reg-istered voters, participated in the elections, electingFrelimo presidential candidate Armando Guebuzawith more than 60 percent support and giving Frelimoa majority of seats in the National Assembly.

9.The delegation leadership was particularlyconcerned about the issue of transparency inthe provincial and national tabulation

process. The Center attempted to observe and assess asmuch of the verification process as possible but washindered by a lack of cooperation by the CNE.Concerns included: questionable tally sheet results,mismatched numbers of polling stations and tallysheets, and a constant mistrust between political partyrepresentatives at the STAE. No provincial electoralcommission released clear information on the finalvoter list, and election authorities were unable toexplain adequately why software problems resulted inthe generation of extra tally sheets. As noted by theConstitutional Council, the election results did notinclude a detailed district-by-district map for everyprovince as stated by law, and the CNE has poorlyexplained the reasons for rejected, stolen, or missingtally sheets. Despite CNE President Arão Litsure’sassurance to Jimmy Carter that observers would beable to review rejected tally sheets and the record ofthe reasons for their rejection, the Center and otherobservers were never able to view such a record of the rejection of 699 presidential and 731 legislativetally sheets.

THE CARTER CENTER


13

10.The Center’s confidence in the overallelection result is based in part on thesuccessful conduct of Mozambique’s

first national parallel vote tabulation. Given the con-troversial 1999 results process and the subsequentpolitical deadlock in party relations, the PVT providedan impartial verification of the official results andserved as a possible conflict prevention tool. TheElectoral Observatory, an umbrella group that encom-passed a number of different nonpartisan domesticobserver groups, conducted the PVT with the Center’ssupport. The availability of the PVT results soon afterthe close of polls allowed The Carter Center to makemore informed judgments about the overall picture.Mozambique’s domestic election observers are congrat-ulated for the conduct of this crucial check on theofficial results.

11.The Center assessed political finance inMozambique to better understand theoperation of political parties, especially

during election campaigns. Campaigns tended to relyon a combination of the party’s share of public funds,free advertisement on public radio and television, eco-nomic investments, and private donors. Frelimo is themost successful party to foster economic activities andis the only party with broad access to private financialsupporters. Personal wealth and party membership feesplayed only minor roles in campaign finance. Despiteclear legal political funding regulations, many aspectsof party finance lack transparency and oversight.Access to state resources is open to a range of abusesthat benefit the ruling party and contribute to unequalparty capacities.

12.Several political parties submitted com-plaints to the Constitutional Council,each of which was denied based on tech-

nical problems with the submission process. TheCarter Center welcomes the strong recommendationsmade by the Constitutional Council, including thegeneration of a single, national voter register; the prop-er institutional and professional development of theCNE; a more consistent knowledge of the electoral leg-islation on the part of the political parties; and theneed to create adequate conditions for electoral obser-vation. The Center is concerned that some issues didnot receive sufficient attention from the council,including the abuse of public resources by political par-ties during the campaign period and acceptance of thedelayed results from the electoral bodies with no suchflexibility being accorded to political parties with lateelection petitions.

Mrs. Carter meets STAE Director-General Antonio Carrascoas CNE President Arão Litsure and Vice President AngelicaSalamão look on. The delegation co-leader, former Presidentof Benin Nicéphore Soglo, is behind Mrs. Carter.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


14

Mozambique’s experience of democratic elec-tions has been marked by the twinchallenges of establishing peace after

decades of conflict and crafting economic develop-ment policies that meet the basic needs of the people.

While other European nations granted independ-ence to their colonial possessions in the years afterWorld War II, Portugal maintained its rule overMozambique for nearly three subsequent decades. In1962, several anti-colonial political groups establishedthe Frente da Libertacã o de Mocambique (LiberationFront of Mozambique, Frelimo). An armed campaignagainst Portuguese rule formed two years later, andafter a decade of intermittent fighting, Portugal relin-quished its colonial rule, granting Mozambiqueindependence on June 25, 1975.

Following independence, Frelimo benefited fromwidespread support as the liberation party. Leaders ofthe Frelimo military campaign established a one-partystate, and in 1977, the party formally declared itselfMarxist Leninist. Frelimo enjoyed complete control ofthe state for two decades during which it expanded its

organizational structure and developed itself at boththe national and local levels. Yet while systematic privi-lege was accorded to the urban, industrialized areas,rural development foundered.

Civil war emerged in the decade following inde-pendence, with the government unable to exert controlover much of the rural areas. In these neglected areas ofcentral Mozambique, the governments of Rhodesia andapartheid South Africa supported the creation of anarmed resistance movement, Resistência NacionalMocambicana (Mozambique National Resistance, orRenamo). Following Zimbabwean independence in1980, South Africa became Renamo’s chief supporter,and the conflict intensified with Renamo becoming themain challenger to Frelimo’s authority.

After President Samora Machel’s mysterious planecrash in 1986, his successor, Joaquim Chissano, initiatedpeace talks with Renamo. The resulting constitution in1990 proposed fundamental shifts in the character ofthe Mozambican state and economy, establishing amultiparty system, market-based economy, and theframework for free elections. The civil war formally concluded with the signing of the Rome General PeaceAccords in 1992. The conflict had killed more than 1million Mozambicans, forced 1.7 million people to takerefuge in neighboring states, and left several millionindividuals internally displaced. Mozambique’s socialand economic infrastructure was devastated, and politi-cal mechanisms for fostering dialogue and managingthe country’s future were nonexistent.

Following the peace accords, Renamo received US$17 million from a United Nations trust fund to helptransform the rebel movement into a political party.Despite generous funding, the party’s transition intothe multiparty political structure was arduous, as mostof Renamo’s leaders had military backgrounds but littleformal education or political experience. The partyalso lacked any substantive ideology besides opposition

A Maputo mural depicts the Portuguese colonial conflictwith Mozambicans.

GR

AN

TL

EE

NE

UB

ER

G

HISTORICAL BACKGROUND

THE CARTER CENTER


15

to Frelimo and suffered from rigid centralization with-in its party structures.

Mozambique held its first democratic elections in1994. A newly formed Comissão Nacional de Eleições(National Elections Commission, or CNE), composedof members chosen on party lines, supervised the con-duct of the elections, technically administered by theSecretariado Técnico de Administração Eleitoral(Technical Secretariat for Electoral Administration, orSTAE). Although not without controversy, the elec-tions were widely determined to be free, fair, andsuccessful, and Joaquim Chissano was elected presi-dent with 53 percent of the vote. His party also won amajority of the 250 seats in the National Assembly,while Renamo gained 112 seats.

While both of the main political parties adaptedto the new multiparty system, they also continued todispute the basic rules of the game, often causing elec-toral system reform to be slow and contentious. Thepromise of decentralization and local democracythrough the 1998 elections in 33 newly created munici-palities was overshadowed by a Renamo boycott. Theparty withdrew from the elections when it felt that itscomplaints about flaws in the registration process were

unanswered. Voter turnout was less than 15 percent,and Frelimo won in every municipality.

The electoral law was rewritten in time for nation-al elections in 1999, and a new voter registration driveprovided voter registration cards to 85 percent of thepotential electorate (more than 7 million voters). Onceagain, Joaquim Chissano was elected president, andDhlakama suffered his second defeat, this time by amargin of less than 4 percent. Overall, compared torecent experiences of post-transition second electionsin Africa, The Carter Center observers foundMozambique’s 1999 general elections showed signs of amaturing political system. The parties forged a consen-sus on the electoral law and campaigned widely, andelection day processes were well-administered with highvoter turnout. However, given the highly politicizednature of election administration and the close presi-dential race, technical problems and a lack oftransparency during the national vote tabulationundermined the credibility of the results.

The Center offered a number of recommendationsfor possible steps to improve future elections that serveas important benchmarks for assessment of the 2004elections. These recommendations included: 1) reform-ing the electoral law to eliminate gaps andcontradictions; 2) restructuring the CNE on the basis ofa comprehensive review involving civil society, politicalparties, and election technicians; 3) increasing the roleof civil society leaders in the CNE, selected in consulta-tion with the political parties; 4) clarifying CNE rulesand operating procedures; 5) restructuring STAE as anindependent body with permanent technical staff; 6)adjusting a series of election day procedures; 7) publish-ing complete polling station results for the 1999elections and for future elections; 8) establishing a fasterreporting system and allowing party agents andobservers to monitor the data; 9) permitting a greaterrole for civil society, media, and national observers togather information about election results, including con-ducting parallel vote tabulations (PVT); 10) adoptingregulations to provide automatically for a review ofresults, or a whole/partial recount, if certain margins or

Two informal market sellers with their homegrown potatoessit in contrast to the availability of Coca-Cola.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


16

thresholds are crossed; and 11) reforming the institu-tions and processes for electoral dispute resolution.

In anticipation of the 2004 elections, The CarterCenter observed Mozambique’s second local electionsin November 2003. New legislation for municipal elec-tions was approved in October 2002, but the CNEremained highly politicized with 10 members nominat-ed by Frelimo, eight by Renamo-UE, and onerepresentative from civil society as chairperson. CNEand STAE structures at provincial and district levelswere similarly composed.

An update of the voters roll, the first since 1999,took place between June 26 and July 26, 2003. Somelogistical problems were noted, although this time,contrary to 1998, the voters roll itself was not contest-ed, and Renamo-UE participated in the elections.

The municipal elections also provided a criticalwindow of opportunity in which to strengthen localobserver groups in advance of the 2004 national elec-tions and to assess whether the recent round ofelectoral reforms was adequate. While voter turnoutincreased to nearly 25 percent, continuing problemswith the voter register remained, and thefinal results process reinforced theCenter’s view that more far-reaching elec-toral reform was necessary.

One important and promising reformwas the establishment of theConstitutional Council. Although estab-lished less than a month before themunicipal elections, its mandate enabledthe council to verify the legal demands forcandidates for the presidential elections,to serve as the last body of appeal forcomplaints, and to validate the final elec-tion results.

Frelimo consolidated its hold on astrong majority of municipalities, winningthe seat of municipal president in 28 andabsolute majorities in 29 municipalassemblies, with victories in many areas

that had voted for Renamo-UE in the 1999 generalelections. Renamo-UE won five municipal presidentseats and majorities in four municipal assemblies.Other parties (notably, Party for Democracy andDevelopment, led by former Renamo Secretary-General Raul Domingos) and civic organizationsperformed poorly, winning a total of only 13 seatsacross all municipalities.

The Center also notes the imperatives of economicdevelopment in Mozambique. Mozambique’s 19 mil-lion people are among the poorest in the world, andthe country ranks 171 out of 177 countries on theUnited Nations Human Development Index. A majori-ty of people live on less than $2 a day, and more thanhalf of Mozambique’s population is undernourishedand does not have access to clean water. AsMozambique looks toward the future, it must addressthese issues to complete the transformation into avibrant and responsive democracy with policies thataddress the most immediate needs of the people.

Mozambique’s long coastline provides a supply of fresh fish brought in byhand from many small-scale fishing operations.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


17

Over the last decade, the Center has supportedMozambican development through interna-tional election observation, support for

domestic observers, review of the political financestructure, assistance in the creation of a consensus-building national development strategy, andimprovement of agricultural production. These effortshave brought together the resources of the CarterCenter’s Democracy Program, Global DevelopmentInitiative, Americas Program, and Global 2000.

The Center’s commitment to democratic princi-ples has initiated a longstanding relationship with theMozambican government and electoral authorities.The Center observed Mozambique’s second multipartygeneral elections in 1999, the subsequent electoral lawrevision process, the 2003 municipal elections, the2004 voter registration update, and the December2004 presidential and legislative elections. It is thisextended and diverse involvement that informs thisreport and its recommendations.

In addition to supporting democracy throughelection observation and technical support,The Carter Center has worked with Mozambicansto improve agricultural productionand to build consensus around a nationaldevelopment strategy, Agenda 2025.

1998

2002

1999

2004

2003

2nd General Elections: December 3-5, 1999

3rd General Elections:December 1-2, 2004

Electoral Law Revisions:1999, 2002 (& 2004)

2nd Municipal Elections:November 19, 2003

Voter RegistrationUpdate and ElectoralLaw Reform

1994 1st General Elections: U.N. supported;October 27-29, 1994

Rome Peace Accords: October 4, 1992

FIGURE 1:MOZAMBIQUE

ELECTORAL EVENTS

1992-2004

1st Municipal Elections:Renamo boycott; June 30, 1998

Voter Registration Update: May 1999

THE CARTER CENTER IN MOZAMBIQUE

THE CARTER CENTER


18

The 1999 elections were Mozambique’s secondelections after its transition from civil war. Assuch, these elections were an important test of

Mozambique’s democratization and, more generally, of itstransition from war to peace and national reconciliation.

At the invitation of the CNE, The Carter Centersent a delegation of 13 observers to the 1999 voter reg-istration and concluded that the process waswell-managed and implemented. In October of thesame year, the Center opened a field office in Maputoand deployed 10 medium-term observers to monitorthe campaign and electoral preparations. Prior to theelections, the Center expressed concern over the delaysin the disbursement of campaign funds, serious inci-dents of violence, credible reports of intimidation ofRenamo representatives, and biased media coverage.Despite these setbacks, the Center was impressed withthe breadth of political party campaigning and diversityof views.

Supplementing the long-term observation, TheCarter Center deployed a 50-person delegation co-led byformer U.S. President Jimmy Carter, Mrs. RosalynnCarter, and former President Ketumile Masire ofBotswana. Over the three days of the election (the elec-tions were originally scheduled for Dec. 3 and 4, butbecause of logistical problems in Zambezia province, theCNE extended voting to a third day), The Carter Centerobserved 747 polling stations in all of Mozambique’sprovinces. After voting closed, delegates remaineddeployed to observe ballot counting and tabulation.

The election itself was tranquil, orderly, and effi-cient, with high voter turnout and results that revealeda tightly contested race between two strong parties andcandidates. Observers did note intimidation ofRenamo representatives in one province and, moreconcerning, an undermining of the credibility of theprocess by a series of technical problems that emergedduring the tabulation of votes, which fueled political

suspicions and caused divisions within the CNE.These problems were compounded by a lack of trans-parency during the final stages of tabulation at thenational level preceding the announcement of the offi-cial results and by the limited technical monitoringcapacity of the parties’ agents and representatives.

The Carter Center postelection assessments recom-mended that the country’s electoral institutions takesteps in future elections to build trust, confidence, andcredibility. The Center offered six recommendations asa contribution to the discussions that were takingplace in Mozambique and emphasized the even greaterimportance of transparency and broad participation inthe process from both civil society and political parties.

The Center saw fine-tuning the electoral law as abasic prerequisite to smooth, efficient, cost-effective, andtimely future elections. Vague laws in the 1999 electionscompelled the CNE to make too many policy decisionsand left election officials at provincial and lower levelsprone to making varying interpretations of the law. Amore detailed list of the Center’s suggested reforms canbe found in this report’s section on electoral law.

The Center suggested restructuring the CNE basedon a civil society, election technician, and politicalparty review of the elections’ organization. An initiallist of possible changes included reducing the size ofthe CNE from 17 members to a more workable num-ber, creating mechanisms to make CNE membershipless partisan and more impartial, establishing clearCNE rules and operating procedures to enhance trans-parency, and streamlining the functions of the variouslevels of CNE (and STAE) offices.

The Carter Center noted the need to clarify thelines of authority between the CNE and STAE andsuggested making STAE an independent body withpermanent technical staff that would work both dur-ing and between electoral periods and over which theCNE would provide general policy guidance.

THE 1999 ELECTIONS

THE CARTER CENTER


19

Based on the findingsof the Center’s electiondelegation and on otherreports, The CarterCenter suggested reform-ing a number of electionday procedures. Thereforms included: homog-enizing rules forproduction and use ofparty agent credentials;standardizing polling sta-tion setup, including theorientation of votingbooths; numbering ballotpapers to expedite tabula-tion; and ending the finalday of voting at an earlierhour to provide more day-light during polling station closing and counting. TheCNE considered 500,000 spoiled ballot papers andaccepted one quarter of them. Furthermore, 7 percentof the polling stations were excluded, as the CNEruled that the result sheets contained serious errors.Transparency in this process was limited becauseobservers and journalists could not verify this processat the national level.

To resolve lingering doubts about the election andallow Mozambique to focus on future elections, theCenter suggested that the CNE announce and publishcomplete and detailed results for the 1999 electionsand future elections and consider permitting civil soci-ety groups, national observers, and independent newsmedia to engage in independent parallel vote tabula-tions (PVT) as a means of verifying and enhancingconfidence in official election results.

Finally, the Center recommended that to avoid thedoubts created by the unprocessed tally sheets in 1999,Mozambique should consider adopting electoral regula-tions which would provide automatically for athorough review or re-count of tally sheets (or a re-count of ballots) if certain margins or thresholds are

crossed and ensure that observers have complete accessto such reviews. In addition, the Center suggestedreforming the institutions and processes for electoraldispute resolution. In 1999, the Supreme Court(whose members are appointed by the president)served as the electoral tribunal in lieu of theConstitutional Council, which is mandated in theConstitution but had not been established. For futureelections, the Constitutional Council should be inposition to fill its constitutional role.

Jimmy Carter speaks to men waiting in line to vote during the 1999 elections .

TH

EC

AR

TE

RC

EN

TE

R

THE CARTER CENTER


20

Mozambique held its second multipartymunicipal elections on Nov. 19, 2003. Theprevious municipal elections in 1998 were

marred by a Renamo boycott, flawed voters roll, andlow voter turnout (less than 15 percent). The 2003 elec-tions were important, therefore, as a measure of theprogress the country had made in the five years sinceits last municipal elections and as an indicator of someof the future issues that could surface in the 2004 presi-dential and legislative elections. These elections werenotable as the first election in which the ConstitutionalCouncil was able to establish its authority and in whichRenamo was able to test its local strength.

In March 2003, Mozambique completed the firstround of significant electoral reforms since the 1999elections. The Center had tracked the irregularprogress of this reform process, which had been fre-quently interrupted by political party posturing withinthe parliamentary commission tasked with making rec-ommendations. By the end of 2002, Mozambique hada new municipal electoral law, and the parties hadreached agreement on the basic structure and decision-making process of the National Elections Commission.

The outcome of the reforms was an 18-personCNE with members drawn from the political partieson the basis of their strength in Parliament. The chairwas selected from civil society representatives, and deci-sions were to be voted in by simple majority. The CNEwould set electoral policy within the framework of theelection law and supervise the work of STAE, whichwould once again manage the actual administration ofthe elections.

In June 2003, the Center accepted the CNE’s invi-tation to observe the municipal elections and opened afield office in October of the same year. For almost amonth, four international observers conducted pre-election assessment in 25 of the 33 municipalities. TheCenter was pleased by the overall positive perceptions

of STAE but noted concerns expressed by members ofboth Frelimo and Renamo in some areas that the par-tisan nature of STAE structures made decision-makingslow. The Center also was concerned by the delayedupdate of the voters roll and inadequate, or uneven,civic education across the municipalities.

In the days immediately preceding the municipalelections, The Carter Center deployed a second groupof observers in seven teams. On election days, theteams visited 60 polling sites and 130 polling tables in11 selected municipalities across six provinces andMaputo City. Observers watched opening of pollingtables, all voting procedures, the closing of pollingtables, the counting of the ballots, and the intermedi-ate tabulation at the level of each municipality.

FRELIMO WINS IN A MAJORITYOF COUNCILS

Frelimo won 28 elections for municipal presidentand 29 absolute majorities in municipal assemblies.Frelimo thus consolidated its hold on a strong majorityof municipalities and won elections in many areas thathad voted for Renamo-UE in the 1999 general elec-tions. Renamo-UE won five elections for municipalpresident and the majority of seats in four municipalassemblies. Renamo-UE therefore had the opportunityto exercise formal executive power in elected office forthe first time in Mozambique in the following munici-palities: Nacala-Porto, Ilha de Moçambique, Angoche,Beira, and Marromeu. However, in Marromeu, theRenamo-UE mayor would have to share power with aFrelimo majority in the municipal assembly. Other par-ties and civic organizations fared very poorly, winninga total of only 13 seats across all municipalities.

CARTER CENTER PROPOSED REFORMSThe Center’s initial statement, released three days

after the election, commended the electoral authoritieson the efficiency and conduct of the elections, the

THE 2003 MUNICIPAL ELECTIONS

THE CARTER CENTER


competence of polling station staff, and the high levelof competition between parties. The generally positivereview of election days also noted low voter turnout.The Center was concerned that some provincial elec-tion officials restricted the mobility of accreditedobservers and in some places, hindered access to theintermediate tabulation process. These prohibitionsultimately precluded the Center from verifying fullythis part of the elections.

Several of the Center’s recommendations focusedon alterations which electoral authorities could maketo the election procedure to facilitate the 2004 general

21

elections. The Carter Center recommended a review ofthe computerized voter register. The voter registerrequired not only correction but also completion andthe integration of the 1999 and 2003 updates to avoidearly controversy during the preparation of the 2004elections. The Center warned that the existing prob-lems of overlapping versions of the roll and multipleand incorrect entries would artificially inflate the vot-ers roll by up to 10 percent, or nearly 1 million voters.

A second concern was the habitually problematicresults tabulation process that was controversial for themunicipal elections in 1998 and 2003 and the nation-al elections in 1999. Specifically, excessive errors intally sheets and nontransparency in the processing oftally sheets at municipal level undermined the credibil-ity and openness of the tabulation of results. TheCenter noted the trend of the electoral process in gen-eral, and tabulation in particular, becoming a two-partybattleground and implored that this trend be reversedto encourage more nonpartisan and transparent elec-tion administration. The Center encouraged the CNEto include other witnesses from other parties and togive independent observers full access to all aspects ofthe tabulation.

Finally, the Center stated that the timely announce-ment of credible results is critical to easing tensions andconsolidating the validity of the election. The delays inresults announcement and the questionable accuracy ofthose results damaged the credibility of the electionprocess. The Center also found errors in the resultsannounced by the CNE on Dec. 4 and in the “correct-ed results” published by the CNE on Dec. 11. TheConstitutional Council had to request the CNE to cor-rect errors and had to pronounce on the complaints.The Constitutional Council published its final rulingon Jan. 15, 2004, confirming the results and annullingall complaints but also criticizing the CNE and theParliament and providing valuable suggestions forfuture improvements to the electoral process. Toaddress these delays, the Center supported theConstitutional Council’s statement that the schedulefor the release of official national results should be

Political differences often lead to violence in Mozambique.Carter Center observer Ricardo de Rodrigues stands in theburned-out shell of Renamo district party offices.

MA

RG

OT

GO

UL

D

THE CARTER CENTER


Mozambique, such as the publication of detailed officialinformation on the delimitation of each municipality.

In addition to making recommendations to elec-toral authorities, the Center made recommendationsto other groups whose participation in elections wasvital to ensuring democracy. These included:

Civil society organizations should be able to collab-orate with electoral authorities to ensure maximumsuccess and coverage of civic education campaigns.

Media workers should receive additional trainingon the electoral process and law and on how best toimprove their coverage of election campaigns withoutpolitical bias.

Political parties should review their internal proce-dures to ensure that internal party democracy prevailsin the selection of candidates for the party lists.

Particular effort to ensure the participation ofwomen and youth, as is the case with Frelimo, shouldbe undertaken by all parties.

Political parties should produce a code of conductwith electoral authorities to govern the behavior ofpolitical parties and their supporters to ensure a peace-ful and tolerant election campaign in 2004.

22

reviewed to ensure that the CNE has sufficientresources for the timely announcement of final results.

The Center reiterated a recommendation previous-ly made in 1999 that the CNE consider reducing itssize and limiting the role of political party representa-tives among its membership. In addition, both theCNE and STAE were again encouraged to create mech-anisms to reassure Mozambicans that they were actingin an impartial and transparent manner. One suggest-ed method of building this credibility was for the CNEto engage in dialogue with international and domesticobservers to ensure improved geographic mobility andaccess of election observers to all aspects of the elec-toral process, including intermediate and nationaltabulation.

The Center also noted aspects of the electoralprocess that will affect future municipal elections in

These elections were notable as the first election in which the Constitutional Councilwas able to establish its authority.

THE CARTER CENTER


23

Mozambique has updated its electoral law anumber of times since the constitutionalrevisions that instituted a multiparty demo-

cratic system in 1990. The timing and content of theserevisions have been very much tied to the electionschedule and outcomes. The revisions governing the2004 elections do not appear to be based on a strategicvision of what the electoral system should look like;instead, the changes appear to once again have beenmade on an ad-hoc basis, informed by problems dur-ing the 1999 and 2003 elections and the prevailingpolitical relations between Frelimo and Renamo.

The Mozambican political system is a multipartypresidential democracy based on the 1990 constitu-tion. Executive power is vested in a directly electedpresident who appoints and chairs a council of min-isters, including a prime minister. The NationalAssembly has 250 members elected by party list ineach province. Seats are awarded based on percent-age of the vote in each province with the stipulationof a 5 percent national threshold for any party togain representation.

While certainly not without difficulties, underUnited Nations guidance and with the strong presenceof international observers, Mozambique’s first generalelections in 1994 demonstrated the country’s potentialfor peaceful democracy and the parties’ willingness tomake concessions to work within this new legal frame-work. Joaquim Chissano was elected president with 53percent of the vote, and Frelimo won 129 seats in theNational Assembly, followed by Renamo with 112 andnine representatives from the Democratic Union (UD).

By 1998, however, optimism waned as the firstlocal elections were plagued with delays and, finally, aRenamo boycott. Recognizing the potential for a down-ward spiral, the National Assembly ratified a newelectoral law (Law No. 3/99) prior to the 1999 generalelections. In 1999, President Chissano was re-elected

by a margin of only 4 percentage points over Renamocandidate Afonso Dhlakama. Frelimo increased itsmajority in the National Assembly to 133 seats, fol-lowed by Renamo with 116 seats and one independent.

Following the 1999 elections, the Center recom-mended important changes in electoral legislation toaddress the irregularities observed by the Center’sinternational election observation delegation.Proposed changes in the electoral law included: 1) clar-ify the eligibility of registered voters who are 18 by thetime of the election although not by the end of theregistration period; 2) review the system of campaignfinance; 3) assure access to all aspects of the electoralprocess for national and international observers; 4)limit voting to a single day; 5) allow individual pollingstations that genuinely need an additional day of vot-ing due to logistical problems to be granted the extraday without burdening the rest of the country with thecost of a third day of nationwide voting; 6) expeditethe counting and tabulation process by, in part, allow-ing decisions about null votes to be made in thepolling stations; and 7) elongate the election prepara-tion period.

INNOVATION, CHALLENGES REMAINIn 2002, Mozambique reviewed the municipal elec-

tion law and the laws governing the composition ofthe CNE (Law No. 18/2002, 19/2002, and 20/2002),and by June 2004, for the third time since 1999, theNational Assembly adopted a new general election law(Law No. 7/2004). There were four main areas ofreform: structure of the CNE, election procedures, tab-ulation of results, and observer roles. The CarterCenter found that despite some productive revisions,many of the concerns highlighted after the 1999 elec-tions were likely to arise again.

For example, in contrast to the Carter Center sug-gestion to reduce the size of the CNE, it was increasedfrom 17 to 19 members, including a president appointed

ELECTORAL REFORM

THE CARTER CENTER


24

by the president of the republic (though nominated bycivil society). The basis for the makeup of the CNEalso remained partisan and tied to the proportion ofseats each party holds in the National Assembly.Ideally, CNE decisions are to be taken by consensus,but Frelimo continued to hold a majority in the CNE,and on more than one occasion, Renamo boycotted

CNE meetings when they objected to the direction ofelectoral policy.

A number of election procedures affecting the pre-election period were altered, including a guarantee topolitical parties that they could check proofs of the bal-lots before printing, a ban on party use of the centralgovernment’s property or goods for campaigning, the

FIGURE 2: SUMMARY OF ELECTORAL REFORMS

PRINCIPAL INNOVATIONS

� Improved definition of institutional roles, inclusion of glossary of electoral terms,and clearer explanation of polling procedures

� Party symbol to be added on the ballot paper of presidential candidates� Formal preview of sample ballot paper for lists and candidates� Sequential numbering of ballot papers� Five new electoral crimes introduced covering: campaign ethics, neutrality and

impartiality, use of public goods, refusal to distribute tally sheets to party agents,obstruction of CNE or STAE staff

� Voting rights for polling staff, police, and journalists (but not valid for partymonitors and national observers) away from place of registration

� Inclusion of a calculator in materials supplied to each polling station� Distribution of copies of tally sheets and minutes to party agents� Tabulation of polling station results by district� The possibility to extend voting for a third day has been eliminated

REMAINING PROBLEMS

� No legal obligation for STAE and CNE to publish detailed polling station resultsand a full report of the elections within a specific time frame

� No provision to ensure the timely and transparent preparation, testing, andverification of national tabulation software

� Tabulation time frames (seven days to produce district and provincial tallies and15 days to produce national results) remain unchanged despite ConstitutionalCouncil recommendation

� Although many registration books contain more than 1,000 names, the law statesthat each polling station should have a maximum of 1,000 voters

� Polls still close at 6:00 p.m., so counting will still take place in the dark

THE CARTER CENTER


25

clarification of qualifications of polling station officials(such as that at least two of the five must speak thelocal language), the prohibition of opinion polls fromthe start of the campaign period until after the declara-tion of results, and a clarification that delgados decandaturas (party agents) would receive their creden-tials from district election commissions.

In addition, the 2004 laws affected election dayprocedures, allowing polling station officials to vote atany station, thereby enabling officials deployed to sta-tions away from their normal place of residence to casttheir ballot; specifying acceptable activity and behaviorwithin 300 meters of a polling station (notably, novoter may say for whom he or she voted or will vote.within 300 meters of a polling station, no observer orjournalist may speak to voters within this area, nor canany campaign materials come within this area), andprohibiting a third day of voting. An amendment inthe 2002 election law improved accountability in thecounting process through the provision that all partydelegates were to be given a copy of the edital (officialtally sheet) and acta (formal minutes of the presidingofficers).

The electoral law revisions also sought to improvethe accuracy and speed of the counting process. First,the laws stipulated that the number of votes gained byeach candidate must be written in both figures andwords to prevent mistakes made by tired polling staff.Second, party agent copies of actas and editais can beused for the count in the case of missing documents atthe provincial and national counts. Also, polling sta-tion staff was to be equipped with calculators toexpedite counting. According to the electoral law, com-puterized results need only be available on a districtbasis, rather than complete individual polling stationresults as recommended by the Center.

In October 2004, the CNE ruled that observerswould have full access to monitor the voting andcounting processes at polling stations, to attain copiesof the result sheets, and to accompany transport ofelectoral material from the polling station to the dis-trict capital and then to the provincial capital. For the

first time, the CNE welcomed observers to organize anindependent parallel vote count and to compare theirresults to the official CNE numbers (described inmore detail below). More troubling, decisions oninvalid ballots–the crux of concerns with the 1999presidential results– would continue to be made dur-ing CNE meetings closed to observers.

With the strong exceptions of inadequate observeraccess to the tabulation process and CNE reclassifica-tion of invalid ballots, the Center optimisticallyreceived many of these reforms but noted that noother adjustments were made in counting proceduresto tackle the fundamental problems that caused errorsand delays in the vote count. Changes in the design ofeditais to prevent simple errors caused by tiredness ofpolling staff or mechanisms for re-counts were notintroduced by the CNE. The tabulation time frames(seven days to produce district and provincial editaisand 15 days to produce national results) were notamended, although the Constitutional Council recom-mended doing so. Also of concern, the method forallocating the number of parliamentary seats to eachprovince was not changed and remains to date arith-metically wrong (there are supposed to be 250 seats,but when the method was used in 1999, it led to theallocation of 251 seats, and one seat had to be arbitrar-ily taken away from one province to bring the numberback down to 250).

The Center believes that through a renewed strate-gic review of the electoral law, further improvements toMozambique’s electoral process could be introduced.The experience of three general elections, the establish-ment of vigorous multiparty competition, theemergence of increasingly active civil society and mediasectors, and the provision of significant financial andtechnical support from the international communityprovide sufficient grounds for a review. Moreover,Mozambique’s neighbors in the southern Africa regionprovide a rich set of responses to the challenges ofpeaceful and credible electoral processes.

THE CARTER CENTER


26

Voter registration in Mozambique is regulatedby Law No. 18/2002 and is to be conductedby STAE on an annual basis. In practice, the

voter register has only been updated before each elec-tion in 1999, 2003, and 2004.

Mozambican citizens, country residents, and citi-zens living abroad who are aged 18 years and older bythe date of the election are eligible to register as voters.The prescribed annual updating of the voter list is con-ducted by establishing stations throughout thecountry. Typically, voter registration stations are estab-lished in local government or other public buildingssuch as schools and, where possible, they are to coin-cide with the eventual location of polling stations. Inthe voter registration station, officials verify and recordeach voter’s personal information to determine eligibil-ity, take a photograph, and issue a voter card. Thevoter registration staff (or brigades) consists of threemembers (supervisor, clerk, and photographer) at least18 years of age. Voter registration brigade members arerecruited through applications from the general public.

The CNE must inform the public of the voter reg-istration update at least 30 days in advance, though apublic information campaign and the results of theupdate must be published at least 55 days before theelection. The results are posted at each voter registra-tion station for a period of 10 days. After some initialdelays in the electoral calendar, the CNE established atwo-week period for the 2004 voter register update,June 28-July 15.

In order to see the voter registration process first-hand, The Carter Center sent a small team of observersto visit more than 150 voter registration brigades innine provinces. The Center’s observers assessed the reg-istration process with respect to several criteria,including the organization of the registration posts, theconduct of the registration officials, the presence ofparty agents and observers, and citizens’ overall assess-

ment of the process. Observers visited several postseach day throughout the registration update period andrecorded their observation for each station. In additionto observing activities at registration posts, the delega-tion met with election officials, local observers, andcivil society leaders throughout Mozambique. Followingthe completion of the process, the Center produced aninterim public report and distributed it widely inMozambique, and it remains available on the Center’sWeb site.

SUMMARY FINDINGSThe Mozambican voter registration update may be

considered in terms of two distinct tracks. The first isthe series of activities that will be called the manualvoter registration update. These activities include theregistration of new voters (new registrations) and voterswho have changed their place of residence (transferregistrations) and the reissue of voter cards to previous-ly registered voters (reissued registrations). The secondis the public verification of the computerized registra-tion details in the 1999 and 2003 books. This process,which will be called the computer voter registrationupdate, is not explicitly mentioned in the law. TheCarter Center’s assessment of the success of these twotracks differed.

Manual voter registration update: The Center foundthat procedures for the 2004 registration of new vot-ers, transfer records, and the reissue of replacementvoter cards were fair and generally implemented in auniform manner. Registration officials were generallywell-trained and well-equipped to perform their task.The pace of individual voter registration was alsonotable, with a national average of 10 minutes toprocess each voter. Carter Center observers noted, gen-erally speaking, a friendly relationship between officialsand party agents. National observers were found atposts in Inhambane, Sofala, Manica, and Tete, andnotably in nearly half of the visited posts, Carter

VOTER REGISTRATION UPDATE

THE CARTER CENTER


27

A newly registered voter displays his voter card.

GR

AN

TL

EE

NE

UB

ER

G

Center observers counted two or three women offi-cials, and only a handful were comprised of exclusivelymale staff.

Stations did face some technical difficulties. Forexample, some brigades ran out of materials, andSTAE regularly suffered from a lack of vehicles.However, the Center did not observe cases where vot-ers could not register as a result of these difficulties. Inmost cases, party agents from Frelimo and Renamo-UEwere present.

The Center did note a discrepancy between urbansouthern brigades and northern rural brigades, withthe former registering nearly twice the number of vot-ers per day as the latter, even though each brigade isresponsible for approximately the same number of vot-ers. The discrepancy was even higher in the central

provinces where urban brigades registered voters at adaily rate four times that of those in rural areas.Notably, the urban centers in southern Mozambiqueare Frelimo. The Center does not believe this correla-tion indicated partisan or fraudulent intent, althoughRenamo would later cite it as an example of bias. It isalso important to note that regional variations in theintensity of civic education before the voter registerupdate and longer distances to registration brigades inrural areas are possible contributing factors in thesedifferent rates of registration.

The computerized voter registration books: The publicverification of the voter register is an important meansof allowing voters to identify and correct entry errors,but this measure is only successful if voters understandits importance and are afforded ample opportunity todo so. The Center welcomed the CNE and STAE

THE CARTER CENTER


28

based on the addition of the updated registrations in2003 and 2004 to the roll from 1999 minus thenational Mozambican annual mortality rate of 1.5 per-cent for people age 18 years and older. TheMozambique National Institute for Statistics (INE) esti-mates that 9,511,001 Mozambicans are eligible voters.In other terms, 1999 + 2003 + 2004 registered voters–1.5 percent of eligible voters = projected registered vot-ers for 2004 (8,559,620).

The registered voter total therefore indicated animpressive registration rate of 90 percent of the totalestimated eligible voter population. However, this fig-ure is higher than the official registration rate in 1999and 2003 and may indicate that many of the approxi-mately 8.5 million registered voters were double entriesor that the names of dead voters remained on the reg-ister. Again, this does not necessarily indicate afraudulent intent to inflate the voter register, as it mayhave been an attempt to ensure that as many eligiblevoters as possible would be able to vote in the 2004elections.

RECOMMENDATIONSOver the course of 2003-2004, the Center strongly

encouraged Mozambique’s electoral authorities to con-tinue improving the voter register to avoid inflation ofregistration data and disenfranchisement of eligiblevoters and to improve the transparency of the voterregistration process by publicizing the steps undertakento give voters the opportunity to check their names onthe register.

Electoral authorities were further encouraged todedicate adequate resources to conduct the following:1) a thorough cross-check of all voter registration

implementation of this procedure and encouragedthem to make this measure standard practice in futurevoter registration updates.

The Center remained concerned about the credi-bility of the computerized voter roll. The Centerobserved only a limited number of voters reviewingtheir inscription data, and registration books wereeither not available for examination (15 percent of thevisited posts) or were incomplete (10 to 15 percent ofvisited posts) for a total of 26 cases (with remarkablylower incidences in the southern provinces). In someinstances, the books also appeared to have databaseerrors such as the repetition of birth dates. The Centerestimated that if these problems in the accuracy of thevoter register were not fixed, legally eligible and regis-tered voters could be refused a ballot paper athundreds of polling stations.

RESULTS OF MANUAL VOTERREGISTRATION UPDATE

Throughout the voter registration update process,1,245,809 people were registered. Of that number,approximately 56 percent were new registrants, 17.5percent were transfer registrations, and 26.5 percentwere reissued registrations. The preliminary STAEresults indicated the 2004 voter registration update waslargely consistent with the update conducted beforethe 2003 municipal elections. The trends in transfersand reissued registrations also appeared to haveremained stable between 2003 and 2004, with themain exception of Sofala, which doubled the numberof reissued registrations from 5.2 percent of the totalin 2003 to 10.4 percent in 2004.

Election authorities failed to make available a finalcopy of the national voter register with breakdowns byprovince, district, and polling station. For this reason,in some cases, it is not possible to determine to whatextent previous recommendations made by the Centerand others were taken into consideration.

On the basis of the available data, the Center pro-jected an estimated 8,559,620 voters were registered toparticipate in the 2004 elections. This projection was

Election authorities failed to make availablea final copy of the national voter registerwith breakdowns by province, district, andpolling station.

THE CARTER CENTER


29

brigade supervisor reports, 2) a review of the manualvoter registration books, 3) an integration of the multi-ple voter registration databases, 4) a sample-based auditof the voter register entries, 5) the implementation ofthe previously announced distribution of the voter rollto political parties, and 6) the publication of the crite-ria or formula used to determine the number ofbrigades per province (e.g. estimate of new voters, pop-ulation density, transport requirements, andinfrastructure, etc).

The Center recognizes that a full comparison of allentries in the computerized books with the originalinformation is a costly and labor-intensive exercise.The Center recommended a sample-based audit: eitheran area-based comparison, referencing voter registra-tion trends since 1994, or a sample-based audit ofinformation from the database, checked against theinformation of selected voter cards. Despite positivereactions to these proposals, with ample support ofcivil society organizations, political parties, and theSTAE director-general himself, an independent auditnever took place.

In the proposed cross-checks, the Center urgedSTAE to focus on the elimination of double registrantsin the computerized voter roll as the number of votersper province determines the proportional distributionof seats in Parliament. The Center did not receive anyevidence that STAE undertook sufficient steps toensure a credible and correct voter register.

The Center also recommended closer institutionalcollaboration to reconcile differences between NationalInstitute for Statistics population projections andSTAE estimates of the voting population as the INEwas found to have the most accurate statistics on voterpopulation and registration in Mozambique. Finally,the Center recommended that STAE analyze demo-graphic data to adjust registration post distribution aswell as civic education strategies.

ROLE OF ELECTION OBSERVERSConsistent with international practice, The Carter

Center firmly believes in the importance of a crediblevoter register as the basis of equitable access to the vot-ing process and electoral choices that reflect the will ofthe people. Observation of voter registration is there-fore a legitimate activity, and appropriate access shouldbe accorded to international and domestic observers.

In addition to direct observation of the 2004 voterregistration update, The Carter Center sought to usethe period as an opportunity to strengthen collabora-tion with domestic observers in overall support oftheir efforts and preparations for the 2004 elections.Despite positive relationships with many key electoralauthorities, the Center found a need for greater under-standing of the role of international election observers(including pre- and postelection observation) in manydistricts. Although properly accredited by the nationalelection authorities, Center observers often werereceived with distrust and insecurity on the part oflocal electoral officials. Indeed, many district and localofficials were unaware of the CNE regulations regard-ing international and domestic election observers. TheCenter proposed that these regulations and moreinformation on the rights and responsibilities of elec-tion observers should be communicated morethoroughly during the training of election officials andsupplemented with the proper distribution of CNEobservation regulations to district and local officials.

THE CARTER CENTER


30

In order to understand the role of money and poli-tics in Mozambique, The Carter Center sent apolitical finance specialist to conduct interviews

with political party representatives; the CNE; and rep-resentatives of the international community, media,civil society, and social scientists. The interviewsfocused on the sources for political finance and mainelements of expenditure, with questions on the inde-pendence of oversight bodies, the role of donors, andthe relevance of the issue in national politics.Additional research based on legislative analysis andnewspaper reports supplemented these interviews. Thecomplete report, from which highlights are includedhere, is available on the Center’s Web site.

WHAT THE LAW SAYSThe Rome peace protocols and the 1991 party law

established basic regulations about political fundingthat remain today (See Table 1). Mozambique’s lawcontains few prohibitions on sources of funding anddoes not limit private donations or campaign spend-ing. This openness to possible donors applies equallyto campaign finance and funds for party development.The law identifies a range of public subsidies, includ-ing public funding, free access to public radio andtelevision in election periods, and tax exemption forthe parties. Although it does not specify the amountof public resources to be allocated, there are clearrules on its distribution. While campaigns receiveequal support, past electoral success determines partyaccess to funds.

The peace agreement also included financial trans-parency as a necessary requirement for party activity.Despite its clarity, this rule has yet to be implemented,and the funding of parties is inadequately documented.While the CNE is expected to provide oversight, itscomposition of party representatives, based upon legisla-tive representation, means the majority of members arefrom the governing party. Although impartial oversight

is possible, any decision favoring the governing partycould be received with skepticism, particularly where thevote is split along party lines.

Mozambique’s electoral code permits sanctionssuch as imprisonment or fines, but they have not beenapplied.

PRIVATE FUNDING PRACTICESSelf-finance: Both the social structure and the insti-

tutional environment hinder the extensive use ofcandidates’ own wealth for politics. Although itreversed the nationalization of private property startingin the 1990s, Mozambique remains a very poor coun-try where business is dependent on the state. Thecountry lacks a large, powerful, and thriving economicclass with a vocation to engage in politics. The use ofpersonal wealth is limited in legislative elections sincecandidates run on a closed party list and parties, ratherthan candidates, are the focus of the campaign.Presidential candidates have more incentive to investpersonal wealth. In the 2004 elections, businessmanand Frelimo presidential candidate Armando Guebuzawas able to self-finance his campaign. However, sincehis party had substantial access to other resources, theGuebuza campaign did not need a large share of per-sonal funds.

The lack of transparent decision-making during the composition and ranking of the Renamo party listcaused internal frustration and speculation aboutwhether money had influenced nominations, butobservers suggested personal networks were likely moreinfluential. For its part, when forming party lists,Frelimo examines a variety of criteria beyond the poten-tial candidate’s capacity to self-fund the campaign.

Membership fees: Frelimo is the only party to raise aconsiderable amount of funds from its base of sup-porters. Frelimo has evolved from a cadre organizationinto a mass party. As a result, the number of partymembers increased from 100,000 in the late 1980s to

CAMPAIGN FINANCE

THE CARTER CENTER


31

2 million members today. The party has also made asystematic effort to raise funds from party members,and membership fees now fund ordinary costs of theparty. Membership dues are calculated in relation toworkdays, averaging about 5 percent of salary. Manystate employees are Frelimo members, thus guarantee-ing Frelimo some public sector funds. Annualmembership fees from all members are estimated to be $1.5 million.

Membership fees play a minor role in oppositionparty funding because partisan hiring practices meanfew opposition members obtain the relatively well-paidpublic sector jobs. Compared to Frelimo, Renamo’smembers pay a monthly fee of just 1,000 meticais, or50 cents a year. Unconfirmed reports indicate Renamocharges more significant amounts of money from elect-ed deputies in the legislature. Given the extensivepoverty in Mozambique and the limited size of partymemberships, smaller political parties cannot count ongrass-roots funding.

Economic investments: Frelimo is the sole organiza-tion with incipient experience in fostering economicactivities as a means of party finance. Numeroussources report Frelimo holds real estate. Party head-quarters in Maputo are in a modern building, which isrented out to private tenants and to a café. Experiencein the city of Beira, where Renamo won the municipalelections in 2003, corroborates this practice. WhenFrelimo yielded administrative control of the munici-pality, the party argued that a number of buildingsused by local government were owned by Frelimo andcharged the new municipal administration with thecosts for renting them.

Renamo also owns real estate with party headquar-ters in the capital and provinces. Part of this capitalstems from the trust fund responsible for financingRenamo’s transformation from a movement of armedrebels to a political party in the early 1990s. The partyheadquarters in Maputo have modest facilities, arelocated in a middle-class neighborhood, and are insharp contrast with the impressive Frelimo building.There are no signs Renamo or any other party is able

to raise funds from other economic activities.Private donations: Frelimo is the only party with

broad access to private sponsors. This access is linkedto the special protection the state–under Frelimo lead-ership–can give to private investors. Because largeparts of Mozambique’s formal economy depend ongovernment for licenses, tax exemptions, subsidies, orcredits, the government’s discretionary power deter-mines the success or failure of an investment.Although such transactions are not documented, it iswidely assumed the state selectively offers credits toparty members and friends of the party. Party mem-bers owning profitable economic undertakings areoften willing to let Frelimo share in their success,which some Mozambicans view as a form of “payback.”

During an electoral campaign, the business com-munity is invited to contribute to the governing party.Urban fund-raising events are disguised as invitationsto discuss economic policies, while in the countryside,businesspersons are solicited for contributions.Refusing these invitations reportedly produces negativeconsequences for business. Experts said corporatedonors to Renamo commit economic suicide, but in

A Frelimo campaign truck, flags, and supporters in party T-shirts illustrate the strength of the party’s resources.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


32

TABLE 1: LEGAL PROVISIONS ON POLITICAL FINANCE IN MOZAMBIQUE

Party finance Campaign finance

Legal sources Party Law (Art.17) enumerates legalsources of party funding: membershipfees, donations and legacies, and publicfunding. A fourth category of “othersources” is undefined.

Electoral Law defines legal sources of party funding: candidates andparties who own resources; donations from national or foreign parties,national or foreign nongovernmental organizations; citizens andforeigners; campaign fund-raising activities; and public funding.

Vetoes, limits,and ceilings

Rome Protocols 1992 do not mentionprohibitions.

Party Law (Art. 19) prohibits donationsfrom state institutions (except officialfunding), from foreign governments, andfrom private institutions of public utility.

Electoral Law (Art. 35, 40) prohibits contributions from state institutionsand foreign governments.

Contrary to party law, private institutions of public utility are not bannedfrom contribution. In fact, NGOs are explicitly included above as a legalsource.

The sanction for undue use of public facilities (Art. 186) is a fine andimprisonment up to one year. No sanctions are provided in case of illegalfunding.

Direct publicsubsidies

Rome Protocols mention direct budgetsupport.

Party Law (Art. 17) also refers to regularsupport with public funds, based onproportional representation in theNational Assembly.

Rome Protocols mention support for parties, based on number ofcandidates.

Electoral Law (Art. 36) states distribution of funds for legislative electionhas to take into account criteria of representation in Assembly andnumber of candidates of each party.

This implies that resources for presidential election have to be allocatedequally among all running candidates.

Indirect publicsupport

Rome Protocols mention fair access tomedia and tax exemption.

Party Law (Art. 15) exempts partiesfrom customs duties and other taxes on anumber of economic transactions.These exemptions do not apply to profit-making activities of the parties.

Rome Protocols propose access to public broadcasting systems should befree of charge.

Electoral Law (Art. 29) states presidential candidates and parties areentitled to use public broadcasting systems (television and radio), leavingthe regulation of access to the CNE.

Accountabilityandtransparency

Rome Protocols mention transparencyrequirement.

Party Law (Art. 16, 18-21) obligesparties to report annually on partyaccounts, including detailed informationon funding sources and disbursements;to publish the accounts in the officialgazette; to register all donations andlegacies; and to hold internal records onall properties.

Electoral Law (Art.37-39) states candidates and parties have to renderaccounts on all revenues and expenditures within 60 days of the officialannouncement of the election result.

After receiving these accounts, the CNE rules within another 60 days ontheir regularity and publishes its decision in the official gazette.

Oversight andsanctions

Oversight is unclear.

Party Law (Art. 20) states that partyaccountability is ruled by the norms ofgovernment accountability. Thissuggests parties have to render accountsto the office in charge of thegovernment audit (TribunalAdministrativo).

Electoral Law states the CNE is the central oversight institution for allissues linked to elections. However parties can appeal CNE rulings withthe Constitutional Council.

Sanctions are usually limited to small fines and in some cases includeimprisonment.

Failure to hold records on campaign finance is charged with a fine.Failure to render accounts may impede right to participate in the nextelection. Cases of parties or candidates with irregular accounts arereviewed by the attorney general. Nonapproval of accounts does notappear to result in docking of public resources.

THE CARTER CENTER


33

reality, this is unclear. Renamo has traditionallyreceived donations from private investors and mer-chants who oppose the Frelimo government, notablythe Portuguese business community and supporters inneighboring African countries.

Among the smaller parties, the Party for Peace,Democracy and Development was a new party led byRaul Domingos, whose experience and internationalcontacts helped him organize and capitalize the PDDcampaign. One source reported the PDD rapidlystrengthened its party organization and infrastructureby building on the social network of a former non-governmental organization founded by Domingos andalso by acquiring real estate for party headquarters inthe provinces. Sources also suggested the PDD receiveda donation of more than $2 million from a foreignbank, but without full disclosure, this cannot be con-firmed. Another small party, the Independent Party ofMozambique, reportedly received support from busi-ness leaders linked to the national and internationalMuslim community.

PUBLIC FUNDING PRACTICESDirect budget subsidies: Public support to parties

includes direct resource allocation and indirect sup-port. Direct annual budget support allows parties toengage in ongoing party-building activities. Althoughthe amount is not defined by law, there are rules con-cerning the allocation of funds to parties: Only partiesrepresented in the National Assembly are entitled tosupport, and funds are divided in accordance with theproportion of seats each party holds in the NationalAssembly. Frelimo and Renamo receive funds from thestate budget amounting to approximately $1.5 millionper year, per party.

During an election year, parties receive additionalfunds. The total amount of public resources for the 2004elections was $2.2 million from a combined source ofinternational donors and the Mozambique government’snational budget. As specified by law, the bulk of publicfunds for campaigns is allocated in equitable terms:

■ One-third of the total amount was reserved for

the presidential campaign (each of the five parties witha presidential candidate received $150,000).

■ One-third goes to parties with candidates run-ning in legislative elections. Since parties may fieldcandidates in only some of the electoral races, theamount of funds is proportional to the number of can-didates the party runs.

■ The last third follows the distribution criteria forannual public resources: Only parties represented inthe National Assembly are entitled, and the allocationof funds follows the proportion of seats (only Frelimoand Renamo-UE qualify for these funds).

Indirect public funding: There are two indirect formsof public support to parties and campaigns inMozambique: free space for advertisement on publicradio and television and tax exemptions. Media spaceis allocated equally to each party, and parties with pres-idential candidates each receive five minutes of airtimeper day, as do parties with legislative candidates.Unlike direct financial support, the number of candi-dates nominated by each party does not temper equity,and even small parties who run in only a few racesreceive equal space on radio and television.

However, free media is a double-edged sword. Onthe one hand, free access to media ads is an importantresource. On the other, the making of television andradio ads is very expensive. Only Frelimo, Renamo,and PDD are able to produce television spots of highquality. Both Frelimo and Renamo hired professionalservices to produce these pieces and update the spotsdaily. Other parties either produce poor television adsor are unable to use the space at all. Two weeks beforethe elections, six parties had not yet presented anyfootage to fill their free ad space on television.

The use of personal wealth is limited in legislative elections since candidates run ona closed party list and parties, rather thancandidates, are the focus of the campaign.

THE CARTER CENTER


34

By law, free media access is limited to the publictelevision and radio stations. Although the law allowsfor additional ads on private stations, parties typicallydo not buy extra time since electronic and printmedia do not have far-reaching coverage outside themain cities.

Another variant of indirect public support is taxexemption. One argument to justify this special treat-ment lies in the noneconomic character of partyorganizations. Another is that taxation and the oversightactivities it requires could eventually result in unduecontrol of political organizations. Tax exemption ismeant to preclude parties from this kind of censorship.

IMPROPER FUNDING PRACTICESAbuse of state resources: There are four types of alle-

gations concerning improper funding in Mozambique.First, the governing party is accused of using stateresources for its own interests. This may include useof public vehicles during the campaign, expensive trav-els of public authorities to distant places with acampaign agenda, and other campaign benefits. Oneexample was a so-called goodbye trip of PresidentChissano to all provinces, where he presented his can-didate, Guebuza, as a natural successor. In addition,Carter Center electoral observers reported severalcases where official vehicles were employed in partyrallies and convoys. The use of state resources bymembers of the ruling Frelimo party appears to be adeeply embedded practice.

Fraudulent use of tax exemption: Tax exemption abuseis reportedly a major source of improper revenue forsome parties. Since many economic activities arelinked to import and export transactions, the parties’import duty exemption makes them a potential vehicle

for tax-free imports that have no legitimate connectionto the parties’ political function. Several reports onparties’ tax-free import and resale of goods on the localmarket support allegations that parties do sometimesuse their tax-exempt status for purely commercial trans-actions. Frelimo reportedly imported 300 tons of paperfor party use, but the material was later sold on thelocal market by a university paper store. This form of“legalized smuggling” includes manufactured productslike vehicles, paper, building material, and even perish-able goods.

There is also evidence of undue use of customsauthority to favor Frelimo. The Center was told of acase in which a small party tried to import 500 televi-sion sets, but the customs authority denied thisapplication. Another party reported that an importa-tion of construction material worth $600,000 wasrefused. There is broad awareness of the discretionarypower of the state apparatus in either refusing oradmitting applications for imports in the name ofpolitical parties.

Delays in allocation of public subsidies: Parties report-ed delays in the allocation of public funds. It is notclear whether this was due to an overall culture of last-minute solutions or whether specific parties had beentargeted. In any case, small parties are more affected bythese delays since they have no capital stock to pre-finance expenses until the paperwork giving access topublic funding goes through the bureaucracy.

Unbalanced media reports: The abuse of publicmedia to promote government candidates has beendenounced on a number of occasions. Early in thecampaign, broadcast coverage on government-ownedRadio Mozambique and on the national public televi-sion station TVM was biased in favor of thegovernment, and the weekly newspaper Domingo fea-tured Frelimo and its presidential candidatedisproportionately. Later campaign coverage improvedsomewhat. Casual observation showed the three mostimportant presidential candidates were covered in areasonably fair way on public channels. Although someopposition supporters complained that TVM’s cover-

The use of state resources by members of theruling Frelimo party appears to be a deeplyembedded practice.

THE CARTER CENTER


35

age of Frelimo’s past activities was inappropriately pre-sented as news and that the Frelimo presidentialcandidate, Armando Guebuza, was favorably present-ed, no outright omission or entirely unfair treatmentof opposition candidates was documented. Observersreported that campaign coverage by public radio wasmore balanced than television.

WEAK TRANSPARENCYIN POLITICAL FINANCE

Political parties only care about the transparencyrequirements for public funding when they areenforced. While the law imposes strict rules abouttransparency of public funds for parties and cam-paigns, de facto public oversight is weak (see Table 2).There are also different rules for accountability con-cerning public and private funds, and despite the legalprovisions requiring disclosure of public funds, political

parties do not feel obliged to render accounts on pri-vate resources for election campaigns. As far as partyfinance beyond the electoral campaign period is con-cerned, parties do not comply with the legal obligationof rendering annual accounts of public and privatefunding, nor do they register private donations as isrequired by law.

The CNE is the most important oversight body onpolitical finance, but it does not publish the content ofreports delivered by parties to account for theirresources. The public is only aware of whetheraccounts have been accepted or rejected. Internationalinstitutions in Mozambique are concerned about thislack of transparency, but local civil society organizationleaders have not yet focused on this issue, and citizensseem unaware of the limited accounting rendered bypolitical parties receiving public funds.

TABLE 2: RULES ON ACCOUNTABILITY, OVERSIGHT, AND TRANSPARENCY OF POLITICAL FINANCE

Party resources

Rome 1991, Party Law 1991

Campaign resources

Electoral Law 2004

Public funding Annual disclosure of accounts

Detailed information on fundingsources

Subject to government audit rules

Prequalification for public funding

Several installments withintermediate accountability

Subject to oversight by NationalElections Commission

Private funding Annual disclosure of accounts

Detailed information on fundingsources

Individual registration ofdonations necessary

Subject to oversight bygovernment

Accounts on campaign 60 days afterelection

Subject to oversight by NationalElections Commission

THE CARTER CENTER


36

The Carter Center began to prepare its long-termdeployment to observe the electoral processshortly after the voter registration. From the

beginning of the mission, the Center made clear to theCNE that it was committed to observing the entireelectoral process, including the political campaigns,election preparations, polling, counting, and the tabu-lation of results. Access to all elements of the processand mobility for observers were among the mainpoints discussed with CNE.

In September, the Center held a round of meet-ings with electoral authorities, political parties, anddomestic observers to assess the status of the prepara-tions of the general elections. Once again, all politicalparties welcomed the Center’s involvement, and theCNE issued a formal invitation to the Center on Oct.26, 2004 (see p. 67).

In mid-September, the Center visited the foreignvoter registration posts for Mozambican residents inSwaziland and South Africa. Voter registration updateposts at the Mbabane Consulate (Swaziland), ElansrandGold Mine (Carltonville), and Pretoria GeneralConsulate showed very little participation, despiteSTAE’s efforts to promote voter registration for thenewly created two National Assembly seats for overseasvoters. Party agents appeared to be trained adequatelyand informed of registration procedures, and no majorincidents were reported. Opposition parties argued thevoter registration update violated CNE regulations bysetting up registration posts not only in embassies andconsulates but in other nonofficial locations such asthe mines in South Africa (where many migrant work-ers from Mozambique are employed). Nevertheless, theprocess was approved by the CNE, and overall, 46,966Mozambicans were newly registered.

In early October, the Center recruited a team ofnine long-term observers. The team received an inten-sive two-day training at a rural post in the district ofMatutuine, covering observation methodology and

logistics, and a full-day workshop with electoral author-ities, Mozambican academics, domestic observers, andpolitical party representatives.

OBSERVATION METHODOLOGYThe Center’s deployment plan for long-term

observers covered all 11 provinces (including Maputocity) and ensured that more than 50 urban and ruraldistricts were visited.

The long-term observers were deployed to assesscivic education campaigns, to advance election prepara-tions, and to collect logistical information for futuredeployment. On Oct. 17, when the formal electoralcampaigns started, the observers received their formalCNE accreditation. However, based on the advance vis-its throughout Mozambique, the Center noted thatsome parties were clearly campaigning before the legalstarting point established in the electoral law.

The Carter Center’s long-term observation report-ed on five main issues: civic education, politicalenvironment, organization of political parties, role ofcivic organizations, and administration of the electoralprocess. In close collaboration with the ElectoralObservatory (a consortium of civil society organiza-tions and religious groups formed for electionobservation), the Center’s observers gathered relevantinformation on the general status of the preparation ofthe election. Over the course of 2003-2004, the Centerworked closely with the Electoral Observatory, provid-ing technical assistance for the conduct of parallel votetabulation, collaborating on several workshops on elec-tion and democracy issues, and providing training onlong-term election observation.

During the campaign period, observers held meet-ings with a wide range of stakeholders, includingelectoral authorities, political party representatives andcandidates, civil society organizations, journalists, busi-nessmen, and traditional and religious leaders, amongothers. On Nov. 16, the Center issued a statement

PRE-ELECTION OBSERVATION

THE CARTER CENTER


37

announcing that President and Mrs. Carter and for-mer Benin President Nicéphore Soglo would lead a60-person delegation for the December elections.While the general environment was calm, observersfound some signs of intimidation. The electoral codeof conduct signed by all political parties seemed to belittle known and thereby weakly respected by mostpolitical activists.

SUMMARY FINDINGSPeaceful political environment: The Center’s

observers found the overall political environmentgenerally peaceful, and major political parties cam-paigned actively in almost all provinces. Even inhighly contested districts, the Center’s observers sawdoor-to-door campaigning by smaller parties, and theCenter’s observers were generally well-received withthose interviewed willing to speak openly about thechallenges faced in their districts. Most respondentscharacterized the political environment as less tensecompared to previous elections. Indeed, in one dis-trict, the major parties reached an explicit agreementon a campaign time line to schedule their respectiveactivities. Also of importance, women were visibleparticipants in the campaign process.

Intimidation and difficulties campaigning werereported in some districts, and although respondentstended not to describe the incidents as major violence,in some areas political actors felt they were unable toreveal their political views without fear of intimida-tion. Notably, opposition parties described politicalspace as “limited or very limited” in some districts inTete, Gaza, Niassa, and Cabo Delgado. In some ofthose cases, opposition parties reported they encoun-tered difficulties setting up offices and were subjectedto verbal provocation and in some instances, stabbingsand assaults. In a series of incidents on MozambiqueIsland, off the coast of the northern province ofNampula, supporters from Frelimo and Renamo threwstones at one another, engaged in a standoff whentheir respective motorcades refused to let the otherpass, and exchanged accusations of defamation.

Early in the campaign period, political rhetoricthreatened to spiral out of control when Renamoleader Afonso Dhlakama threatened that the countrywould slide back into war unless the people voted forthe opposition. Frelimo candidate Armando Guebuzareplied that such politicians “are lying, because theycannot wage war in Mozambique. The war is over, andthis is not the time for more violence.”

All political parties signed an electoral code of con-duct on May 14, 2004. Articles in the code included acommitment to resolve electoral conflicts through dia-logue, cooperate with the electoral authorities, andrenounce the use of false or defamatory allegations,intimidation tactics, and bribery. Although it held nolegal weight, if effectively publicized, such codes holdthe potential to delineate effective ethical parametersand can encourage more tolerant political discourse andbehavior. Unfortunately, the Center’s observers foundthat the code was not properly distributed among par-ties and, therefore, was applied only inconsistently.

In a Nov. 22 press release, the Electoral Observatoryexpressed serious concern with the conduct of politicalparties during the campaign. They singled out the use ofyouth to destabilize their opponents’ campaign events,in contravention of the code of conduct, as well as

FIGURE 3: MEMBERS OF THE ELECTORAL

OBSERVATORY

� Christian Council� Islamic Council� Catholic Bishops Conference� Human Rights League (LDH)� Mozambican Association for the

Development of Democracy (AMODE)� Center for Studies in Democracy and

Development (CEDE)� Organization for Conflict Resolution

(OREC)

THE CARTER CENTER


38

PROFILES OF PARTIES 2004PARLIAMENTARY ELECTIONS

(In order of the official ballot paper)1. Renamo-Electoral Union is a coalition

between Renamo and 10 other opposition parties.As in 1999, Renamo reserved two places for each ofthe parties. In 1999, Renamo-UE received 38.79percent of the vote for Parliament. For 2004,Renamo’s unchallenged leader and presidential can-didate was Afonso Dhlakama (see profiles ofpresidential candidates, page 41).

2. Frelimo (Mozambican Liberation Front) isthe former liberation movement in charge of gov-ernment since 1975. President Joaquim Chissano,party leader since 1986, announced after the 1999elections that he would not run in 2004. In 2002,the party congress elected Armando Guebuza secre-tary-general and Frelimo’s presidential candidate(see profiles of presidential candidates).

The following parties did not meet the thresh-old of 5 percent of the popular vote to gainparliamentary representation:

3. PDD (Party for Peace, Development andDemocracy) is led by Raul Domingos, former chiefof staff of the Renamo army in the civil war andhead of the negotiation team during the 1992 peaceagreement with the government. Domingos ledRenamo’s parliamentary wing from 1994 until his2000 expulsion from the party by Dhlakama.Although some expected PDD to attract supportfrom frustrated Renamo members, the party failedto pass the 5 percent threshold.

4. PARENA (National Reconciliation Party) ranin all 11 provincial constituencies. PARENA is run-ning for the first time.

5. SOL (Social-Liberal Party) ran in all 11provincial constituencies. The SOL won 1.7 percentin 1994 and 2 percent in 1999.

6. Ecological Party-Land Movement ran in all

11 provincial constituencies. It is one ofMozambique’s three ecological parties but lacks avisible ecological message.

7. PIMO (Independent Party of Mozambique)ran in all 11 provincial constituencies. Of the estab-lished minor parties, PIMO is one of the few thatproved able to win in the November 2003 munici-pal elections when it won three seats in municipalassemblies (in the northern towns of Nampula,Angoche, and Cuamba). PIMO won 1.2 percent in1994 and 0.71 percent in 1999.

8. PASOMO (Social Broadening Party ofMozambique) ran in all 11 provincial constituen-cies. The party won only 0.05 percent of the vote in1999.

9. PVM (Green Party of Mozambique) ran inall 11 provincial constituencies. PVM is one ofMozambique’s three ecological parties and does nothave any electoral record apart from a 0.27 percentshowing in the 2003 election for the Maputomunicipal assembly.

10. PAREDE (Democratic Reconciliation Party)ran in all 11 provincial constituencies.

11. PT (Labour Party) ran in six provinces. PTstood in previous elections, taking 0.6 percent in1994 and 2.7 percent in 1999. It stood in severalmunicipalities in the 2003 municipal elections butdid not win a single seat.

12. PPD (Popular Democratic Party) ran candi-dates in Maputo city and the two diasporaconstituencies.

13. FAO (Broad Opposition Front) is a coali-tion between the FL (Liberal Front) and the PAC(African Conservative Party). FAO ran only inInhambane and Maputo city.

14. MBG (United Front for Change and GoodGovernance). MBG is a coalition between UNAMO(Mozambique National Union) and PARTONAMO(Party of All Mozambican Nationalists) and ran in

THE CARTER CENTER


39

insults and intimidation in campaign rhetoric. The use of and access to public resources were

additional flashpoints during the campaigns.Opposition parties regularly complained of Frelimouse of state resources for party purposes (e.g. use of

state vehicles for party campaign, Frelimo campaignposters and flags posted to public buildings), unequalopposition-party access to public spaces for campaign-ing, as well as unequal public media coverage of thedifferent party campaigns.

On Sept. 23, the CNE adopted a code of conductfor political parties, candidates, and other politicalgroups (Deliberation 34 of 2004). While this code wascomprehensive in its description of prohibited behav-ior, it did not establish a clear monitoring andmediation responsibility in the event of infractions.

Perhaps of greatest concern is the sense that whilethe general political atmosphere was calm, thereremains an underlying current of tension and suspi-cion in Mozambique’s politics. To be sure, Frelimo hasby far the best-organized party structures, evident inthe quality of their campaign materials and rally organ-ization, and so it follows that they are able to appealmore effectively and persuasively to the people.However, Frelimo exercises its influence in many sub-tle, and often institutionalized, expressions as the partyin power. For example, a local public official makes anadministrative decision based on a partisan interpreta-tion of the law and receives backing from the police,or a civil servant campaigns openly for the oppositionand receives a warning to stop or face transfer to a lessdesirable district. Thus, the public institutional space,not just the physical world of buildings and resourcessuch as vehicles but of political regulation and behav-ior, tends to be dominated by one party. Theconsequence for Mozambique’s political developmentis that even if authorities conduct an election thatmore or less complies with the procedures required bylaw or accepted international practices, this electoraldemocracy does not necessarily guarantee the demo-cratic transformation of the state and politics. Thechallenge of building a democratic politics is more dif-ficult to assess since it requires a keen understandingof context, language, culture, and the many subjectivecomponents that enable Mozambicans to make moralsense of their world.

Role of observers: On occasion, election observers

all provinces except Inhambane and Maputo city.15. CDU (United Congress of Democrats)

broke away from PALMO after the 1999 electionsand appeared only on Zambezia’s ballot paper.

16. PAZS (Party of Solidarity and Freedom)stood in all provinces except Cabo Delgado, Tete,Manica, and Gaza. This party ran for the first timein 2004.

17. UD (Democratic Union) ran in all provincesexcept Niassa and Cabo Delgado. In 1994, a coali-tion called the Democratic Union (UD), formed bythe Liberal and Democratic Party (PALMO), theNational Democratic Party (PANADE), and theMozambican National Party (PANAMO), took 5.15percent of the vote and won nine seats. After someparties broke with UD before the 1999 elections, itsshare of the vote declined to 1.5 percent, and it lostall of its seats.

18. PALMO (Liberal and Democratic Party ofMozambique) ran only in Niassa, Cabo Delgado,Nampula, Gaza, and Maputo provinces. In 1999,PALMO broke with the UD and won 2.5 percent.

19. USAMO (Union for the Salvation ofMozambique). A coalition between PADRES(Democratic Alliance for Social Restoration), PSM(Socialist Party of Mozambique), PSDM (SocialDemocratic Party of Mozambique), and UM (Unionfor Change), USAMO ran in all provinces exceptInhambane, Gaza, and Maputo.

20. PADELIMO (Democratic Liberal Party ofMozambique) ran for all provinces except Nampula,Zambezia, Tete, Sofala, and Inhambane. PADE-LIMO won 0.8 percent of the vote in 1999.

RIGHT: Frelimo campaignposters promote the party, outgoing President JoaquimChissano, and presidentialcandidate Armando Guebuza.

THE CARTER CENTER


40

can find themselves caught up in the dynamics of thepolitical situation. One such incident occurred in Teteprovince when the provincial Elections Commission(through the local press) accused the Center’s long-term observers of partisan behavior by conducting ameeting with Renamo activists inside an electoralbuilding. In fact, Renamo had no formal representa-tion or offices at the district level, and the meetingtook place with the mutual agreement of both parties(Frelimo and Renamo) represented on the electoralcommission. Indeed, the CNE’s directive of Nov. 18explicitly stated that observers may be present at meet-ings and campaign activities in closed or restrictedareas. This issue was later clarified with the provincialauthorities and closed without further complaint, butit illustrated the importance of good communicationregarding the independence and neutrality of theCenter’s role as election observers, especially whenpolitical competition is intense.

Election preparations: The Center’s observers report-ed that the training for polling officials proceededwithout major incident, although Renamo representa-tives in some provinces claimed that the recruitmentprocess was not fair or transparently conducted. Themost consistent complaint of STAE officials was inade-quate transport. Overall, the Center’s observersreported general public confidence in STAE efforts toensure that the supply of election equipment, materials,and staffing were on track for the Dec. 1-2 elections.

One notable oversight was that despite the closeresult of the presidential election in 1999 and theuncertain level of support for a credible third candi-date, Raul Domingos, STAE, and the CNE did notundertake any planning for a possible second round.In the event that no candidate secures more than 50percent of the vote, the electoral law calls for a secondround between the top two candidates to take placewithin 21 days of the announcement of results fromthe first round.

Inadequate voter register: Sufficient information wasnot always forthcoming from the CNE. The CNE’srepeated reluctance to provide a detailed list of polling

PH

OT

OS:

GR

AN

TL

EE

NE

UB

ER

G

ABOVE: The former Renamopolitician, Raul Domingos,stood as presidential candidatefor his newly formed Party forPeace, Democracy andDevelopment.

BELOW: Third-time Renamopresidential candidate AfonsoDhlakama was the main challenger to Frelimo in 2004.

THE CARTER CENTER


41

Armando Guebuza was born on Jan. 20, 1943.During the years prior to independence, Guebuzawas involved in active guerrilla fighting, rising to therank of general. Since Mozambique’s independence,he has occupied high-level government positionssuch as minister of interior and vice minister ofdefense, among others.

During the collapse of the Soviet Union,Guebuza was one of the first Mozambican leaders torecognize the need to establish a multiparty marketsystem and to create links with U.S.-based interna-tional financial institutions such as the World Bankand the International Monetary Fund. Guebuzabenefited from the privatization of state-owned com-panies and is considered one of the wealthiestMozambican citizens.

He enjoyed very high approval ratings from theparty’s congress for his presidential candidacy, andhe was described as a hard-liner and a nationalistwith a populist authoritarian style. His preferencefor nationalist economic policies has raised someconcerns among the international community. Atthe same time, he is viewed as a reformist who canimplement effective policies to address problemssuch as crime and corruption.

Raul Manuel Domingos, presidential candidatefor the Party for Peace, Development andDemocracy, was born Oct. 14, 1957, in Teteprovince. Renamo’s most popular figure after itsleader Afonso Dhlakama, he won renown first as amilitary strategist and then as a skilled politicalnegotiator.

Captured in 1980 by Renamo, he joined theguerrilla movement where he was soon noticed forhis talents as a military strategist and given military

promotion until, in 1987, he won promotion asRenamo chief of staff. He relinquished his militaryleadership in 1989 and became head of the ForeignRelations Department.

In 1991, Domingos led the movement’s delega-tion to the Rome peace talks, and his display ofsuccessful negotiating skills contributed to his por-trayal as Dhlakama’s rival. In the October 1994parliamentary elections, Raul Domingos was electedMP for Sofala province and was appointed presidentof Renamo’s parliamentary group. In 2000,Dhlakama expelled Domingos from Renamo, as hesaw his success and rising popularity as a threat forleadership within the party.

Afonso Macacho Marceta Dhlakama was bornJan. 1, 1953, in Sofala province. He was called toserve in the Portuguese colonial army and fought inthe infantry until 1974, when he joined the Frelimoforces battling for independence.

During the transitional government set up inSeptember 1974 prior to independence until 1977,Dhlakama was head of the logistics department forgovernment troops in Sofala province. It was duringthis posting that he met the future leader ofRenamo, André Matsangaisse, and came to share agrowing opposition to what they considered to bethe authoritarian leanings of the new regime. InOctober 1979, Matsangaisse was killed, andDhlakama took over the leadership, becoming presi-dent and commander-in-chief of the Renamo forces.Dhlakama led the peace talks and was the signatoryof the 1992 General Peace Agreement in Rome.

In the October 1994 presidential race, Dhlakamawon 33.7 percent of the vote as Renamo’s presiden-tial candidate. He ran again in the 1999 electionsand secured 47.7 percent of the vote.

PRESIDENTIAL CANDIDATE PROFILES

THE CARTER CENTER


42

stations with their corresponding register book num-bers foreshadowed a number of impacts for the pollingprocess. Political parties have a direct interest in know-ing where voters are located if they are to get theirmessage out to active and potential supporters. In pre-vious public reports, the Center expressed concernover the geographic distribution of polling stations andthe considerable distances that voters had to travel insome rural areas. Secondly, election authorities canonly plan effectively if they know how many pollingstation kits and ballot papers need to be suppliedbased on the distribution of voters. The ongoing STAEeffort to computerize the voter list and to base a resultsdatabase on a national, consolidated voter list was alsoaffected by the apparent inability to produce the list.Moreover, this information is of public interest.

After the completion of the voter register updatein July, central STAE, in coordination with UnitedNations Development Program technical advisers, con-ducted a consolidation of the electoral register in orderto eliminate duplicate entries, mistakes in spellingnames and birth dates, and other entry errors.Although the status of these efforts was periodicallypresented to observers and the international communi-ty, by the date of the election, the consolidation of thelist was reportedly concluded but not yet systematicallyimplemented and communicated to the provincial elec-tion authorities. On Nov. 4, the CNE published apartial 2004 voter list, but it was not broken down bypolling station. In spite of the insistence on the partof both domestic and international observers, the finallist was never provided by the CNE.

No single voter list consolidating the 1999, 2003,and 2004 voter registers was approved by the CNE,although the director-general of STAE claimed on atleast one occasion that a single computerized voter listexisted. At the same time, one senior Frelimo memberof STAE described the voter register as confidential,and even as classified, state information.

Civic education: The Center’s general assessment ofthe civic education campaign is that it was acceptable,although delays in the distribution of civic education

materials, lack of sufficient equipment, and logisticalconstraints had an impact on its national coverage anddepth, especially in rural areas of the central andnorthern provinces. Given the low voter turnout,which, though varied throughout the country, hoveredaround 40 percent, the real impact of the civic educa-tion methodology remains questionable.

Civil society involvement: In addition to STAE’s civiceducation efforts, the active involvement of some civicorganizations, notably the Civic Education Forum(FECIV), was a positive sign of state-civil society collab-oration. The cooperative network of nongovernmentalorganizations engaged in domestic election observationunder the umbrella of the Electoral Observatory was asecond positive example of the increasing scope ofinvolvement of civil society organizations in the elec-toral process. The Electoral Observatory enableddiverse organizations to pool their resources and initi-ate a more comprehensive election assessment than anyof the individual groups could have done on their own.The Carter Center’s observers maintained a close part-nership with these organizations and their members,both at central and local levels. Carter Center staff wasable to work in a genuinely cooperative effort with theElectoral Observatory, sharing information, providingskills, and building capacity.

Nevertheless, civil society organizations hoping totap into the political space enabled by Mozambique’sturn to multiparty democracy remain highly depend-ent on donor agencies, and in most cases, the groupsremain focused primarily on elections. In addition,some organizations raised concerns over what wasdescribed as a thin delimitation between the preferredneutrality of civic education and the political interestsof some agents. In some cases, the Center’s observersreceived complaints about biased recruitment of theseagents. While it is difficult to state definitively if somecivic organizations were consciously partisan, insuffi-cient supervision, inadequate resources, andtransportation limitations reveal the overall weak struc-tural condition of civil society organizations and thosewho work for them. Elections tend to result in more

THE CARTER CENTER


43

donor funds for civil society groups, but these fundssometimes arrive late in the process, organizations maylack the capacity to spend them effectively when theydo arrive, or their previous access to fewer resourcesimpacts negatively on their ability to sustain reliableprograms with broad geographic coverage.

Media coverage of campaigns: The Center’s pre-elec-tion observers received complaints about biased mediacoverage, particularly from television broadcasterTVM and the newspapers Noticias and Domingos.For example, PDD, the party led by former Renamopolitician Raul Domingos, claimed that TVM failedto transmit coverage of Domingos and showed himonly in appearances with children. Frelimo andRenamo presidential candidates also traded mediaaccusations of rallies attended mostly by children,apparently as an effort to expose the other’s supposedweak support among eligible voters.

Renamo went further and, at one point, bannedTVM cameras from filming party rallies in the towns ofMozambique Island and Nacala. Although proportionalbroadcast time on public radio and television was avail-able to all presidential candidates and parties, thecampaign period was otherwise marked by partiality inmedia coverage. The impact of media bias, especially asit relates to the often illiterate majority of the rural elec-torate, remains unclear, and much more work remainsto be done to fully understand the operation of politi-cal and public communication in Mozambique.Reflecting this relative confusion, the Center receivedmany differing viewpoints regarding the media and per-ceptions of bias in access and coverage.

In one positive sign, the state-owned RadioMozambique, which has national coverage and broad-casts in local languages, proved to be less partial andmore professional than might have been expected.Another positive indicator was that community-basedradio broadcasts were considered informative and edu-cational, and some stations developed a code ofconduct for radio broadcasts during the election peri-od. This evaluation is especially notable, given thatmost community-based radio stations are small, oper-

ate under difficult technical and financial conditions,and still manage to provide an independent voice.Perceptions of print media also covered a spectrum ofopinion, with weekly newspapers such as Savana andZambeze and some fax sheet papers like Mediafax seento be more independent, while others are consideredone-sided, such as Vertical and Imparcial. In any case,the reach of print media beyond the main urban areasis quite limited.

Security forces: The opposition parties consistentlyreported that they felt the police acted in a mannerthat favored Frelimo and its supporters. They citedthat complaints raised by opposition parties did notreceive the same attention as those presented by theruling party, while imprisonment of oppositionactivists was reported at least in various districts inGaza and Tete, Niassa. Reports from the Center’sobservers confirmed some of these claims, particularlyexamples of police intimidation in Frelimo-dominatedareas where opposition parties attempted to campaign.

Collaboration with international representatives: TheCenter maintained close consultation with other inter-national organizations and embassies supporting theelectoral process, namely the European Union, theCommonwealth, United Nations DevelopmentProgramme, and many international missions. TheCenter coordinated its long-term deployment planwith the European Union Observation Mission whilethe UNDP office in Maputo served as an effectivefocal point for coordination and information-sharingamong individual country missions, election authori-ties, and others.

THE CARTER CENTER


44

OBSERVATION METHODOLOGYA fundamental principle of the Carter Center’s

election observation methodology is the understandingthat the electoral process is owned and run by the peo-ple of Mozambique. The guiding principle can be seenin the Center’s efforts to limit an observer’s role toobservation, rather than mediation, and to offer sup-port for domestic rules, electoral institutions, andobserver groups. That said, the Center’s internationalobservation activities can still bolster burgeoningMozambican democracy by allaying concerns anduncertainties about the electoral process, increasingvoters’ and political parties’ confidence in the legitima-cy of the election and Mozambican electoralinstitutions, and reinforcing the work of domesticobserver groups and Mozambican stakeholders.

In keeping with the supportive role in which theCenter sees its work, the delegation was informed thatas election observers, they should avoid involvingthemselves in, or obstructing the conduct of, the polland that they should not comment publicly on thequality of the election process until the delegation as awhole has had the opportunity to debrief.

Upon arrival in their deployment areas, theobservers met with domestic observer groups and heldmeetings with local election officials and representa-tives of political parties to learn about the immediatestate of election preparations as well as the prevailingpolitical dynamics in the area in which they would beobserving the elections.

The work of short-term election observers comple-ments the activities of the long-term observers. Bothtypes of deployment supported the various elements ofthe Center’s program in Mozambique, including obser-vation of election preparations, support to domesticobservers, and the conduct of the parallel vote tabula-tion as well as establishing communication anddialogue with political parties. Whereas the pre-election

The Carter Center organized a delegation of 60observers from 23 countries that arrived onNov. 27 and remained in Mozambique

through the elections until Dec. 5, 2004. Former U.S.President Jimmy Carter, Mrs. Rosalynn Carter, formerBenin President Nicéphore Soglo, and Dr. JohnHardman led the delegation. Over 50 percent of theCarter Center’s observers spoke Portuguese, andalmost an equal number of men and women partici-pated in the delegation.

Observers spent a full day in briefings on Nov. 28prior to their deployment to each of the 11 domesticconstituencies (see pp. 68-69 for map and table ofdeployment). During the days before the elections,observers familiarized themselves with their deploy-ment areas and planned their election day route.They also met with relevant local and provincialauthorities, including the Provincial ElectionsCommissions (CPE), political party leaders, police,and other international and domestic electionobservers. In the meantime, the delegation leadershipheld meetings with national electoral authorities andstakeholders in Maputo.

Several other international observation groupswere present for Mozambique’s elections, including theEuropean Union, Commonwealth, Southern AfricanDevelopment Community Parliamentary Forum,African Union, and the Electoral Institute of SouthernAfrica. The EU had the largest delegation with 130observers, followed by The Carter Center. Where pos-sible, Center staff in Maputo met with theircounterparts in the other organizations to coordinatedeployment and share information. In addition, theprimary grouping of domestic observers, gathered asthe Electoral Observatory, organized a national deploy-ment of observers and conducted a parallel votetabulation (PVT) with technical assistance from TheCarter Center.

ELECTION OBSERVATION

THE CARTER CENTER


Jimmy Carter met with Frelimo presidential candidateArmando Guebuza before the elections.

45

observation focused on preparation for the electionsand the political climate, the short-term observationemphasized the procedural and logistical conduct ofelection day. Short-term observers used checklists torecord detailed information on the area surroundingthe polling station, the people within the polling sta-tion, the operation of the station, the openings andclosings, and other comments on the procedures andevents affecting the voting process.

On Nov. 28, the short-term delegation was briefedon Mozambican politics, elections, and their observerroles and responsibilities. The briefings reviewedMozambique’s election law and procedures, politicalparties, and the code of conduct and observationmethodology. Several political parties participated in around-table discussion, including representatives ofFrelimo, Renamo, PDD, and PIMO. Representativesfrom the CNE and STAE also spoke to the delegationto answer any questions about the electoral process.The delegation also heard from a representative of thedomestic organizations participating in the ElectoralObservatory, who described their program or workand methodology. The Carter Center’s long-term

observers demonstrated the polling process andexplained the use of the Center’s forms to record theirobservations.

DELEGATION LEADERSHIPThe Carter Center observation mission was multi-

layered and inclusive. While observers were deployedacross the country, meeting with local and provincialelectoral authorities, observation leaders met with keystakeholders in Maputo whose decisions have nationalimplications.

The delegation leadership met with all of the majorpresidential candidates, including Frelimo secretary-general and presidential candidate Armando Guebuza,Renamo candidate Alfonso Dhlakama, PPD leaderRaul Domingos, and PIMO’s Ya-Qub Sibindi. MDGcandidate Carlos Reis was unavailable. In addition,delegation leadership team members met with CNEPresident Arão Litsure and other members of the com-mission, STAE Director-General Antonio Carrasco,Mozambique President Joaquim Chissano, andConstitutional Court President Rui Balthazar. TheCenter also met with the leaders of a number of inter-national and domestic observer groups including theEuropean Union, Commonwealth, Brazão Mazula ofthe Center for Studies in Democracy and Development,Otilia Aquino of AMODE, and other members of theElectoral Observatory.

The delegation leadership was pleased to hearstrong endorsements of the democratic process fromall presidential candidates, and they each pledged torespect the election results. The emerging message wasthat Mozambique had turned its back on conflict inorder to face the challenges of economic and socialdevelopment. One major recurrent topic in the leader-ship meetings was the twin issues of proceduralintegrity and political transparency in the tabulationprocess. A lack of definitive information regarding thenumber and distribution of registered voters, the num-ber of polling stations, the quality of the tabulationsoftware and training of data entry technicians, theprocedures for handling problem tally sheets and

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


46

invalid ballots, and continued uncertainty regardingthe access for observers preoccupied and concernedthe delegation leaders. Many of these questionsfocused on the ultimate credibility of Mozambique’selectoral process and whether the country could shedthe suspicions and accusations stemming from 1999.

OBSERVER ACCESSThe Carter Center respects Mozambique’s right to

define regulations concerning election observation.The Center also asserts that, if invited to observe anelection, a credible and useful assessment can only beconducted if observers have sufficient access to the rel-evant aspects of the election process.

In an Oct. 21 press release, the CNE presented fur-ther information on the role of domestic andinternational observers. The Center welcomed theCNE acknowledgment of the importance of nonparti-san observers and their important contribution tocredible, legitimate, fair, and transparent elections.However, a regulation barring observers from speakingto voters within 300 meters of a polling station seemedto be an unnecessarily stringent prohibition. TheCenter understands the importance of noninterferencein the election process, especially inside or in the

immediate vicinity of, polling stations, but it is widelyaccepted that observers may interact discretely withelection officials in order to obtain information rele-vant to the conduct of the poll. However, the CNEfailed to clarify this issue as late as a Nov. 24 informa-tion session for international observers when oneFrelimo official advised observers to greet polling sta-tion officials but avoid any election-related content intheir conversation. The issue was only somewhat clari-fied when CNE Chairman Arão Litsure informed theCenter that observers could pose questions to pollingofficials but should not conduct “interviews” since thiswould distract them from their work.

The CNE confirmed that beyond the polling sta-tions, observers had the right to accompany results tothe district and provincial levels, to receive copies ofthese results, and to conduct their own summation ofresults. But the opportunity to directly observe subse-quent phases in counting and tabulation was heavilyrestricted. Observers were prohibited from circulatinginside the data entry areas at provincial and nationallevels, making it impossible to verify if polling stationtally sheets were entered correctly. However, observers

Carter Center observers traveled by plane and groundtransport to reach their deployment areas. This is professor Scott Taylor and his driver Nuno. E

LM

AD

OE

LM

AN

While most polling stations were located in schools and otherpublic buildings, this thatched hut served in Angoche.

EL

MA

DO

EL

MA

N

THE CARTER CENTER


47

and journalists would be able to observe the data entryarea through a glass partition and to follow results onseparate computer terminals, enabling, in theory,observers to compare results from the tallies postedoutside polling stations with those entered in the data-base. Observers were also prohibited from attendingmeetings of the CNE, in which the political represent-atives of Frelimo and Renamo would review pollingstation tally sheets and invalid ballots.

In a Nov. 22 letter to the CNE, the members of theElectoral Observatory presented a thorough and well-crafted argument on the principle of observer access tothe full electoral process, including tabulation. Theybased their argument on the electoral law, the CNE reg-ulations for observers, and statements of democraticprinciples from Southern Africa DevelopmentCommunity, the African Union, and InternationalIDEA (in all of which Mozambique is a member state).

The Carter Center, EU, and Commonwealthobserver teams shared these concerns regarding observ-er access to all phases of the election process. Whilethe CNE initially allowed direct access only to thecounting at polling stations, they subsequently agreedto allow for limited observer access to the data opera-tions room at provincial and national levels whereresults were tabulated. Although the CNE said thatobservers could follow these results on computerscreens in a separate room, this functioned only imper-fectly in practice, and given the lack of a disclosed listof polling stations, it was impossible to search theresults in a reasonable manner. It was possible to com-pare posted results from outside polling stations withthe corresponding entry in the database but only byscrolling through the entire list of hundreds of tally

sheets. It was impossible therefore to search by pollingstation number or by district. CNE regulations alsoprovided for observers to receive copies of the minutesfrom the operation room, although this did not occur.

The software auditors (a Mozambican firm,Solutions Ltd.) who reviewed the results database madeit clear that they had not evaluated the final tabulationsteps. Given the absence of a final, official list of pollingstations and registered voters, it became effectively impos-sible for observers to know how many polling stationsexist, how many would be included in the final count,and how polling stations with problem tally sheets or ahigh number of invalid ballots would be handled. At theprovincial level, STAE technicians only received andbegan to install the software after the election days, andstaff was subsequently trained at the last minute. Theprovincial data entry was slow and often delayed due totechnical difficulties.

Most voters walked to the polls or rodebikes as at this Gaza polling station.

BR

AD

AU

ST

IN

The Center was particularly impressed withthe careful attention to detail demonstratedby election officials as they administeredpolling procedures.

THE CARTER CENTER


48

The most important stage of the resultsprocess, the CNE sessions to review tallysheets and invalid ballots, was closed toobservers. The CNE claimed that these ses-sions were regular meetings of the electoralbody and, as such, closed to the public. TheCenter raised this issue with the CNE onseveral occasions, most directly whenPresident Carter informed the CNE thatThe Carter Center would be unable to fullycertify an election without access to tabula-tion. In the end, international observerswere invited, on an ad hoc basis and forvery limited time periods, to view the CNEmembers at work, reviewing invalid ballotsin pairs with one representative each fromFrelimo and Renamo.

The point the Center and other inter-national observers tried to make was not tosee the reasons for the reclassification ofeach and every invalid ballot sent to theCNE from the provinces nor to review eachrejected polling station tally sheet. These are amongthe assigned roles of the party representatives on theCNE. Rather, the Center maintained that in order forthe CNE to implement a fully transparent tabulationof official results, the reasons for such decisions shouldbe recorded openly, and this record should be availablefor review. The Carter Center reminded the CNE thatthey did not include nearly 7 percent of polling sta-tions in the final count of 1999 because of problemswith tally sheets. In such cases, the CNE rejectedentire polling station results on the basis of errors orinconsistencies in the sheets.

SUMMARY FINDINGSThe delegation observed almost 1,000 polling sta-

tions over the two days of the election, Dec. 1-2, andreturned the following day for a debriefing and pressconference announcing the Center’s initial assessmentof the election.

On the election days, the short-term observers visited

polling stations throughout the country, noting the com-pliance of the stations with national electoral law.Observers recorded the time polling stations opened andclosed (polls opened at 7:00 a.m. and were required toremain open until 6:00 p.m. with all eligible voters in lineat that time guaranteed the right to vote), whether stationswere free of campaign posters and security personnel,whether all polling station officials were present (pollingstation staff included a president, vice president, secretary,and two clerks) and whether at least two of them spoke alocal language, whether party agents from more than oneparty were present, and how the voting procedures werefollowed. In addition, observers noted the presence ofother international and domestic observers.

Observers remained at one site for a poll closingwhere they noted whether all voters who were in lineat 6:00 p.m. were able to vote, whether the ballotboxes were sealed correctly, and whether police andparty agents were present to guard the boxes overnightas stipulated by electoral law.

These Gaza voters sought shelter from the sun as they waited outside theirpolling station.

BR

AD

AU

ST

IN

THE CARTER CENTER


49

After observing the closing procedure on theevening of Dec. 1, observers called in their first reportsto the Center’s office in Maputo. Observers across thecountry reported a peaceful atmosphere with well-organized voters and voting procedures and low voterturnout. The majority of polling stations opened andclosed on time. In every single observed polling sta-tion, the required number of polling station officialswas present. In Maputo city, at least, the 300-meterrestriction on political posters was nearly universallyignored as billboards and campaign posters from allparties were usually visible from polling stations.However, no complaints on this issue were voiced tothe Center’s observers.

On Dec. 2, Carter Center observers againwatched the opening of polling stations, notingwhether ballot box seals were undamaged, and contin-ued to observe through counting, which in some casesran into the early hours of the morning. Carter Centerobservers witnessed only two cases where serious viola-tions should invalidate results at a polling station. Atone polling station in Quelimane, polling officialsclosed at 1:00 p.m. In Angoche, tensions ran higherthan in other areas when several Renamo party agentswere arrested and stones were thrown, followed bymore arrests the following day. These specific eventswere reportedly indicative of a more generalized intentto intimidate opposition Renamo supporters.Moreover, Renamo party agents there and in somelocations in Cabo Delgado reported they had beenchased away from staying with the ballot boxes (as istheir right) through the night of Dec. 1. In Teteprovince, one of the Center’s teams reported they weremade to feel unwelcome by the CPE and instructednot to speak to polling station officials. However,another of the Center’s teams in a different part ofTete reported no problems. In the overwhelmingmajority of observed stations, however, there were noproblems, and only a small percentage of stations hadminor problems.

Observer records of the second day of the electionwere similar to the first, reporting a largely peaceful

environment, well-prepared election officials, and con-tinued low voter turnout. The Center’s observersreported very positive collaboration and informationsharing with other international and domesticobservers. However, observers were impressed with themeticulous attention to detail on the part of electionofficials and party agents throughout the voting proce-dures and especially during counting at the pollingstations. The delegation leadership, split into twoteams covering Maputo city and Maputo province, con-firmed these assessments.

PRELIMINARY ASSESSMENTThe delegation debriefing on Dec. 3 echoed earlier

reports of orderly voting with a few exceptions.Although international observers are usually relativelyfew in number, a properly implemented election obser-vation methodology that combines long- andshort-term deployment, awareness of election laws andprocedures, and linguistic and country knowledge aswell as careful evaluation of a broad range of informa-tion sources allows observers to offer a meaningfulassessment of the quality of the complex proceduraland political elements of an election. As the table of

The president of the polling station must display and explainthe ballot paper to each voter. The candidate’s name, partylogo, and photograph appeared on the presidential ballot paper.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


50

summary evaluations indicates, when compiled, multi-ple reports of both an objective (e.g. poll opened ontime) and a subjective (e.g. poll functioned well) naturecan provide a composite assessment based on a relative-ly small sample. The Center’s initial assessment of theelections reflected these reports and congratulatedMozambique for the successful conduct of the coun-try’s third multiparty national elections. In a Dec. 4press conference, delegation leader President JimmyCarter noted the elections were generally well-organ-ized: Polling stations functioned effectively, were fullystaffed, and had necessary election materials. TheCenter was particularly impressed with the carefulattention to detail demonstrated by polling station offi-cials as they administered polling procedures (see p. 80for full statement).

Nonetheless, the apparent low voter turnout (esti-mated at the time to be no more than 50 percent andlater determined to be 3.3 million voters or 36 percent)was of concern. Low turnout may be explained in partby the inflated numbers on the voter list, which was

not properly updated prior to the election. PresidentCarter urged election authorities, political parties, andothers to encourage greater political participation andbuild public confidence in the efficacy of the politicalprocess. In addition, he noted that while there were nosystematic problems regarding the voter register appar-ent on election day, the list with the number ofregistered voters per polling station should have beenmade available to political parties and others prior toelection day.

While the initial assessment was mainly positive,the Center also noted several problematic incidents,including several incidents that resulted in the arrestof Renamo party agents and supporters in the city ofAngoche in Nampula province. The Center’s observersalso reported opposition party complaints of policebias in favor of Frelimo during the election campaignand on the election days. President Carter noted hopesfor a more secure, tolerant, and impartial enforcementof the law.

Before receiving a ballot paper, each voter must display thattheir hands are free of indelible ink, applied after they havevoted to prevent double voting.

GR

AN

TL

EE

NE

UB

ER

G

Carter Center observer Marc de Tollenaere, Carter CenterExecutive Director John Hardman, and delegation co-leaderNicéphore Soglo confer outside a Maputo polling location.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


51

The short-term delegation returned from theprovinces to Maputo after the individualpolling stations completed their tabulation.

The Center maintained its long-term team deployed inthe Maputo headquarters of the CNE and STAE andin all provincial capitals except Chimoio in Manicaprovince to monitor the tabulation of the results at aprovincial level. One long-term observer also conduct-ed a round of meetings to assess the politicalenvironment in Beira after the CNE’s announcement.The postelection observation was also conducted inclose consultation with other observers, notably theEuropean Union. In the capital, the Commonwealthexpert team also provided useful information.

COUNTING AND TABULATIONThe counting process began after the polls closed

on Dec. 2 at 7:00 p.m. The polling station president,with other election officials, party agents, and in manycases, national and international observers watching,counted the number of unused ballot papers and thenremoved the seals from the ballot box, opened thefolded ballots, and stated the full name of the candi-date for whom the vote was cast. The clerk recordedthe results on a tally sheet (edital) and completed theminutes of the polling station (acta). All official partiespresent signed these documents. After the ballots werecounted and the tally sheets filled in at each individualpolling station, they were sent to the provincial capital.At the provincial level, there was an intermediate tabu-lation where the contested and invalid ballots wereidentified and sent for reclassification to Maputo. Theelection law states that provincial results must beannounced seven days after the election (Dec. 9).

Not one provincial election commission (CPE)completed this process by the legal deadline. STAE ini-tially blamed the delay on the late arrival of materialsfrom the districts to the provinces because of rain andother transport difficulties, but it became apparent

that the lack of technical preparation was one of themain reasons. By Dec. 6, Sofala, Nampula, and Gazahad not even started their tabulation, while collectionof results was interrupted by rain in several provinces,including Niassa, Cabo Delgado, Tete, and Manica. Insome other provinces, however, including Gaza, Sofala,Tete, and Cabo Delgado, Renamo members on theelectoral bodies kept the warehouse containing elec-toral materials closed and delayed the work ofprovincial election commissions.

Technical difficulties persisted throughout theentire tabulation process. The database at the provin-cial level contained a higher number of tally sheetscompared to the one officially announced and postedby the CNE on Nov. 4. When asked to clarify thesediscrepancies, the CNE contention that improperinstallation of the audited software and insufficienttraining of the technicians were responsible tended toreinforce mistrust of the provincial electoral authori-ties rather than boost confidence in their efforts.

Limited observer access to the tabulation processmade it difficult to generate an independent evalua-tion. The Center’s observers were able to follow thedata entry of the computerized results at STAE head-

SIG

RID

SPIN

NO

X

POSTELECTION OBSERVATION

Carter Center field representative Nicolas Bravo (L) and senior program associate David Pottie (R) thank PresidentCarter prior to his departure from Mozambique.

THE CARTER CENTER


52

quarters, but the software did not enable searches bypolling location. While searching to cross-check resultscollected at the polling stations, the Center’s observersfound tally sheets with unrealistically high voterturnout, including multiple instances of polling sta-tions in Niassa and Tete recording a 100 percentturnout and more than 90 percent support forFrelimo. Given the low turnout nationwide (rangingfrom 30 to 40 percent), ballot boxes appeared to bestuffed at polling stations in the Tete districts ofChangara, Chifunde, and Tsangano as well as in theNiassa districts of Metarica and Marupa and in theGaza district of Chicualacuala. Shortly after the elec-tions, President Carter personally requested of CNEPresident Arão Litsure that observers be present forthe CNE review of individual ballots sent from pollingstations to Maputo. Although the CNE initially grant-ed this request, observer access was subsequentlyrestricted to two hours a day.

As the tabulation process dragged out, the mainpolitical parties began to exchange public accusation.On Dec. 9, Afonso Dhlakama alleged the electionswere fraudulent and called for new elections (to bepaid for by the international community). Frelimoaccused Renamo of obstruction during the tabulationprocess, blaming the party for delays in Gaza, Manica,Cabo Delgado, Niassa, and Maputo. The head of theFrelimo parliamentary group, Manuel Tome, referredto Renamo as a “bad loser” making “ridiculous declara-tions.” Dhlakama would reiterate many of hispreviously stated complaints in a Dec. 22 press release.In the statement, he claimed that a majority of opposi-tion supporters in the central and northern regions ofMozambique were excluded from voter registration,Frelimo forced public employees to support Guebuza’scandidacy, Renamo party agents were chased awayfrom polling stations by Frelimo members, and presi-dents of polling stations received instructions fromFrelimo to stuff ballot boxes in favor of ArmandoGuebuza and to invalidate ballots cast for Renamo.

Also on Dec. 9, domestic observers confirmedreports of some malpractice but praised the overall

conduct of the elections as a significant advance inpolitical rights of Mozambicans. The ElectoralObservatory declared in its preliminary statement“clear and unequivocal satisfaction at the peaceful andorderly way in which Mozambican voters expressedtheir right to vote.” In addition, the observatory notedit was able to gather results data within the parametersof the random sample for its parallel vote tabulation.

On Dec. 17, the day on which the official resultswere due to be announced, STAE Director-GeneralAntonio Carrasco invited observers and journalists toa meeting where a team of technicians offered expla-nations of the discrepancies. The STAE technicianssaid the system produced “fictitious” tally sheets forthose polling stations assigned more than one voterregister book (each book has a maximum of 500names, and each polling station may have up to 1,000voters). STAE acknowledged the problem on Dec. 8and instructed the provinces to ignore the polling sta-tion number generated in the database and use theoriginally (manually) assigned number instead. In thecontext of an otherwise secretive computer section,STAE is to be congratulated for issuing a plausibleexplanation of the existence of such a serious techni-cal mistake that could have constituted fertile groundfor allegations of fraud.

Mozambique’s electoral law calls upon the CNE toannounce the national tabulation of results within 15days after the election (in this case, Dec. 17). However,the CNE announced the final results on Dec. 21, fourdays after the legal date. The Center retained seriousunanswered questions about the complete accuracy ofthe results and the lack of transparency during theCNE’s final tabulation. The official results did notinclude the required district-by-district map of results.

The CNE announced the final results on Dec. 21, four days after the legallyrequired date.

THE CARTER CENTER


53

The results also excluded some 699 presidential tallysheets (5.4 percent of total votes) and 731 legislativeassembly tallies (5.7 percent of total votes) withoutexplanation for their exclusion. There are many rea-sons why a tally sheet cannot be processed, such asfigures that do not add up and cannot be corrected bythe CNE, were unreadable because they were coveredin spilled ink, or were lost or stolen and never reacheddistrict or provincial capitals.

Unfortunately, despite the CNE’s assurance toPresident Carter that observers would receive the CNErecord of reasons for the rejection of tally sheets, norecord was made available, and the Center could notexamine even a sample of rejected tally sheets. Nodetails of the provincial distribution of excluded tallysheets were released either. This rate of exclusion wassimilar to that of the 1999 elections, but the differencein support between the two major presidential candi-dates was much greater in 2004, and, therefore, theoverall result was unaffected.

On Dec. 20, the Center released a statementexpressing many of these concerns, noting the irregu-larities observed during the provincial tabulation, thelimited access for observers to all phases of the count-ing, the delays in the announcement of the results,polling station tally sheets with serious irregularities,software inconsistencies, and political mistrust amongSTAE members.

PETITIONSThe electoral law requires election petitions be filed

with the CNE within two days of the announcement ofthe results. This meant the deadline for petitions wasDec. 23. In the event a complainant wanted to appealthe CNE decision, these had to be filed with theConstitutional Council within five days of the CNE’sdecision. The president of Mozambique appoints thechair of the Constitutional Council, and the politicalparties represented in Parliament nominate the othermembers. Upon appointment, members of the councilmust renounce their party affiliation and may serve upto two five-year terms. The council serves as an appeal

body, but its validation of results is not dependentupon the resolution or absence of appeals.

Several opposition parties filed complaints andappeals, including the Party for Peace, Democracy andDevelopment (PDD), Independent Party of Mozambique(PIMO), the Enlarged Opposition Front (FAO), theMovement for Change and Good Governance (MBG),and the Renamo-EU coalition. PDD wrongly presentedits petition first to the Constitutional Council and wastherefore rejected. PIMO, MGB, and FAO called for theannulment of the elections, but their petitions were alsorejected by the CNE.

On Dec. 27, Renamo presented a petition to theCNE listing the party’s complaints and calling for newelections within six months. The CNE sent the peti-tion directly to the Constitutional Council withoutany deliberation. The council rejected this course andsent the petition back to CNE, demanding formaldeliberation. On Jan. 3, 2005, the CNE rejectedRenamo’s claims, arguing that their complaint was sub-mitted after the legal deadline.

Several of the CNE rulings were appealed and sentto the Constitutional Council for final evaluation. Theappeals of the smaller opposition parties were rejectedon Jan. 12. Renamo presented its appeal to theConstitutional Council on Jan. 10, but the submitteddocument differed in several aspects from the originalpetition sent to the CNE. In this second version,Renamo called for the annulment only of part of theelection. On Jan. 15, the council rejected the Renamoappeal, arguing that it was inconsistent with the origi-nal petition as submitted to CNE, and, therefore, itwas a new petition and could not be ruled upon. Thecouncil also noted that the petition was submittedafter the legal deadline and that the alleged irregulari-ties should have been reported to the district-levelelection authorities at the time they occurred. Theconsequence of these proceedings was that theConstitutional Council ruled out the Renamo appealof the CNE decision on procedural grounds ratherthan through an evaluation of their concerns.

THE CARTER CENTER


54

TABLE 3: OFFICIAL 2004 ELECTION RESULTS

VALIDATED BY THE CONSTITUTIONAL COUNCIL

VOTES % Presidential election

CandidatesRaul Manuel Domingos 85,815 2.73Armando Emilio Guebuza 2,004,226 63.74Jacob Neves Salomao Sibindy 28,656 0.91Afonso Macacho Marceta Dhlakama 998,059 31.74Carlos Alexandre dos Reis 27,412 0.87

Total number of valid votes 3,144,168 99.44Total number of null votes 88,315 2.65Total number of blank votes 96,684 2.91

Legislative electionParties or coalitionsRenamo Electoral Union 905,289 29.73Frelimo 1,889,054 62.03PDD 60,758 2.00PARENA 18,220 0.60SOL 13,915 0.46PEC-MT 12,285 0.40PIMO 17,960 0.59PASOMO 15,740 0.52PVM 9,950 0.33PAREDE 9,026 0.30PT 14,242 0.47PPD 448 0.01FAO 7,591 0.25MBG 11,059 0.36CDU 1,252 0.04PAZS 26,686 0.88UD 10,310 0.34PALMO 9,263 0.30USAMO 8,661 0.29PADELIMO 3,720 0.12

Total number of valid votes 3,045,429 91.68Total number of null votes 109,957 3.31Total number of blank votes 166,540 5.01Total number of votes 3,321,926 100.00

THE CARTER CENTER


55

CONSTITUTIONAL COUNCILREVIEWS RESULTS

In Deliberation 19 of Jan. 19, 2005, theConstitutional Council validated the official results aspresented by the CNE. The Center was concernedthat the council acknowledged important problemsbut did not hold the CNE accountable for the irregu-larities and delays during the tabulation of votes. First,the council noted that the CNE failed to complete afinal tabulation map of district-by-district results forsome provinces; second, the CNE failed to provide arecord of the reasons for the excluded polling stationtally sheets; and third, the council recognized thepotential impact of excluded polling station tally sheetson the final distribution of seats in the NationalAssembly, but no action was prescribed. The final elec-tion results are presented in Table 3.

However, the Constitutional Council voicedimportant criticisms of the conduct of the elections,including recommendations to generate a singlenational voter register, to establish the proper institu-tional and professional development of the CNE, tobuild a more consistent knowledge of the electoral lawsamong members of political parties, and to improvethe conditions for electoral observation.

On Jan. 27, the Center issued a public statementthat congratulated Armando Guebuza for his electionas president of Mozambique. Although the Center rec-ognized this overall result, it also concluded that theCNE did not administer a fair and transparent elec-tion in all parts of Mozambique. The Centerwelcomed the Constitutional Council’s deliberationbut, in addition to the issues noted above, remainedconcerned about a missed opportunity for the councilto comment on other important issues, including thereported abuse of public resources for partisan purpos-es and accepting delayed reporting of election resultsfrom the CNE without equivalent flexibility beingaccorded to political parties during the very tight timelines of the petition process. A particular problem thatwent unmentioned was the reluctance of district-level

electoral commissions to accept party complaints afterthe elections, especially since the council citedRenamo’s failure to submit electoral complaints at thedistrict level as one of the three main reasons forrejecting its petition.

Ultimately, the Center’s confidence in the overallelection results was supported by the independentcheck provided through the parallel vote tabulationconducted by the Electoral Observatory. Additionalresults as published in the media also contributed tothis confidence. The Center agrees with the assess-ment of the EU that while the irregularities duringthe elections and tabulation of results are cause forconcern, they did not affect the overall outcome ofthe election. Noting that the distribution of NationalAssembly seats might have been slightly affected, theCenter concludes that the overall legislative resultsand the election of Armando Guebuza to the office ofthe president clearly reflected the will of theMozambican people.

TABLE 4: DISTRIBUTION OF

PARLIAMENTARY SEATS

Constituency Seats Frelimo Renamo

Niassa 12 09 03Cabo Delgado 22 18 04Nampula 50 27 23Zambezia 48 19 29Tete 18 14 04Manica 14 07 07Sofala 22 06 16Inhambane 16 15 01Gaza 17 17 0Maputo Province 13 12 01Maputo City 16 14 02Africa 01 01 -Overseas 01 01 -Total seats 250 160 90

THE CARTER CENTER


56

attended the training. Participants shared their assess-ments of the voter registration process, noting errors inthe computerized registration books, insufficient cover-age by registration brigades in some areas, theweakness of civic education in rural areas, and prob-lems in receiving accreditation. The seminar alsoreviewed the role of observers and recent reforms inthe electoral law.

■ The Center collaborated with domesticobservers once again in August 2004 to organize a pub-lic seminar assessing the voter registration process.These observations were shared with a range of organi-zations from civil society, the media, representativesfrom political parties, and the electoral authorities.

■ Carter Center staff from Atlanta also visitedMozambique over the course of 2003-2004. Duringthese trips, the Center’s representatives met with vari-ous election stakeholders and reviewed programoperations.

The most intensive and sustained area of collabo-ration between The Carter Center and civil societyorganizations was the provision of technical supportfor the conduct of a national parallel vote tabulationof the December 2004 presidential and NationalAssembly elections. This exercise was a success and rep-resented a long-term commitment of resources andhard work on the part on the Center’s staff inMozambique, with special acknowledgment of the cru-cial role played by the Center’s field representative,Marc de Tollenaere, in support of every aspect of thePVT.

The Carter Center first raised the idea of parallelvote tabulation prior to the 1999 elections. This obser-vation technique was new to Mozambique and wasultimately rejected by the CNE on the basis that suchan initiative was not provided for in the electoral law.However, the controversial tabulation of the 1999 elec-tion results led many to believe that a PVT would havebeen helpful. The Carter Center and others noted with

The Carter Center initiated a number of activi-ties to broaden the role of civil societyorganizations in Mozambique democratic

processes of 2003 and 2004. These activities includedcapacity building, technical assistance and trainingwith domestic observers, fostering civic dialogue withMozambique’s political parties, and sharing experi-ences from previous Carter Center electionobservation missions. The Center’s staff inMozambique met with the entire range of electoralstakeholders in a variety of formal and informal inter-actions in order to achieve the overall goal ofstrengthening the role of civil society in the democraticand electoral process.

EXAMPLES OF SPECIFIC ACTIVITIES:■ In March 2004, the Center, in collaboration

with the Electoral Observatory and AMODE, conduct-ed a seminar on election observation, reviewing thelocal government elections and extracting lessonslearned in anticipation of the upcoming Decemberelections. The seminar participants reflected on thepurpose, performance, achievements, and shortcom-ings of election observation in Mozambique.Participants included electoral authorities, parties,media, nongovernmental organizations, and relevantinternational partners. The Center also presented thefindings from its observation of the 2003 municipalelections in a number of public and private meetingsin order to discuss possible improvements to the elec-tion process.

■ In May 2004, the Center observed a nationalconvention of Mozambique’s political parties thatfocused on the newly adopted political party code ofconduct and also observed a municipal by-election inXai-xai, the capital of Gaza province, on May 19.

■ In July 2004, the Center conducted training onlong-term observation for members of AMODE.Coming from seven different provinces, 16 observers

PARTNERSHIP WITH CIVIL SOCIETY

THE CARTER CENTER


57

concern suspicious modifications to vote tallies andvoter turnout rates at the provincial level and a highrate of errors in tally sheets processed by the CNE inMaputo. Renamo members of the CNE refused to vali-date the official results, and the party submitted aformal petition to the Supreme Court alleging fraudand serious irregularities. Although the SupremeCourt upheld the exclusion of a significant number oftally sheets, it rejected Renamo’s complaint in itsentirety and validated the official results. The followingdescription of the Center’s work with domesticobservers in the conduct of parallel vote tabulationreveals the importance of detailed logistical prepara-tions to ensure the integrity of this observation tool.

THE 2003 PVTIn 2003, the international donor community

worked with civil society organizations to identify will-ing participants to take on the technical and politicalchallenge of a PVT. Efforts were also made to sensitizethe election authorities, party leaders, governmentofficials, the diplomatic community, and the media

about the value of a PVT, andthe new CNE president, ArãoLitsure, supported the conductof a PVT as a means to helpbuild confidence in Mozambiqueelections. The institutional homefor this initiative was a coalitioncomprised of four organizations(AMODE, CEDE, the ChristianCouncil, and the IslamicCouncil). The idea of a PVT wasstill quite controversial, especiallyamong political leaders, andsome civil society organizationswith important electoral experi-ence declined to participate.

Having decided that a PVTin all 33 municipalities holdingelections on Nov. 19 would notbe feasible or necessary, 10

municipalities were selected on the basis of criteriasuch as the prevailing political situation, the expectedcloseness of results, and geographical location. Threecities–Maputo, Beira, and Nampula–were chosen fora PVT by random sample of polling station results,and comprehensive PVT was conducted in seven othermunicipalities.

At the central level, the Electoral Observatory des-ignated a senior PVT adviser, a PVT coordinator, anda small technical staff to support the database andcomputer system necessary for the collection and proc-essing of the results. At the municipal level, a networkof 10 coordinators was selected from the participatingorganizations. In turn, 441 local observers were recruit-ed for observation of the polling and collection of theresults. The Carter Center provided logistical adviceand technical assistance throughout the design processand trained the municipal coordinators, who in turntrained the individual poll watchers.

Forms for a qualitative assessment of the ballotingand counting process were prepared as well as resultstally sheets for the observers. Additional verification

Domestic observers watched thecounting process and reported theresults to Maputo for the parallelvote tabulation.

SIG

RID

SPIN

NO

X

THE CARTER CENTER


58

forms were developed to enable the municipal coordi-nators to verify the data provided by the observers aswell as generic correction forms as necessary. The com-munication system was designed to allow the PVTobservers to call in the results directly to the centralPVT office (except for Maputo) and to deliver theirtally sheets to the municipal coordinators who forward-ed them to Maputo for further analysis. Carter Centerelection observers (14 total) were also deployed mostlyto municipalities included in the PVT to provide anadditional check on the validity of the informationprovided by the domestic observers.

Generally, data arrived at the central PVT officemore slowly than planned, mainly due to a unilateraldecision by the municipal coordinators to collect theresults from their observers instead of allowing themto call Maputo directly. The data entry process was alsoslowed by unanticipated problems with the databaseand an insufficient number of data operators to main-tain full staffing. Despite these difficulties, within 24hours of the polls closing, complete preliminary PVTresults for eight municipalities were available, followedsoon after by results from Beira. In Nampula, the pollwatchers failed to call in the results directly from thepolling stations as previously agreed, and the samplepoints were not always respected. Nevertheless, thesereports were verified and corrected against the resultsfrom The Carter Center and other observers.

In nine of the 10 elections where a PVT was held,the results were very close to the official resultsannounced by the municipal election authorities. ForMaputo and Nampula, the results of the PVT werewithin a 2 percent margin of error. For Beira, however,

the PVT results fell outside the desired statistical mar-gin. In the case of Marromeu, the PVT resultsindicated a victory of the Renamo candidate while theprovisional official results gave the victory to theFrelimo candidate. During the verification of officialresults, the CNE reversed the result in favor of theRenamo candidate (as indicated by the PVT).

Based on this experience, domestic observers andthe Center learned several important lessons. Theplanned PVT for the 2004 elections would benefitfrom the database construction carried out in 2003,but more time and resources would need to be invest-ed to verify some important inputs such as lists ofpolling stations and candidate lists. Second, theobserver training had to be revised with greater empha-sis on the proper implementation of thecommunication system in order to speed up the proc-essing and analysis of the PVT data.

THE 2004 PVTThe 2004 elections presented several significantly

more challenging problems. First, the logistics of trans-port, data collection, and communication were morecomplex because many observers would have to oper-ate from rural areas, which was not the case in 2003.Second, the design of a PVT for the legislative elec-tions required separate samples for each provincebecause representatives in the National Assembly areelected on a provincial basis. Third, given the closepresidential race in 1999, the political stakes werepotentially high for 2004, and it was expected that thepolitical parties would watch the PVT implementationvery closely.

The preparatory activities for the PVT consisted ofestablishing a reliable and feasible sample methodolo-gy; the identification of the coordinators, supervisors,and observers; the formulation of an implementationplan; the design of the observation and tabulationforms; and a deployment plan. Various samplingmethodologies were tested using a database with the1999 election results. Normally, a computer draws arandom sample, but in this case, repeated efforts did

The supervisors and coordinators wereresponsible for the identification and trainingof nearly 1,600 observers, preparation of adeployment plan, and the practical imple-mentation of the PVT.

THE CARTER CENTER


59

not render a consistently low margin of error.One problem was that the election authority was

unlikely to publish a list with the ID number of eachpolling station. Such a list would have enabled thesample to be random to the level of individual pollingstations, but the CNE only published a list of pollinglocations with the total number of polling stations ineach location. In addition to considering a fair geo-graphical distribution, the sample also needed tobalance 1999, 2003, and 2004 polling stations andtheir accompanying segments of the voter register. Afinal challenge was posed by the sharp increase in thetotal number of polling stations from 8,000 in 1999 to13,000 in 2004. This increase was not unevenly distrib-uted over the various provinces nor was it justified byan increased number of registered voters.

In the end, tests were conducted to produce a sam-ple based on the likely list of polling locations ratherthan individual polling stations, and it was found that,

although not ideal, this sample would yield reliableresults. An eventual sample of 792 polling locationswas adopted. This sample was higher than statisticallyrequired, but it took into account the required “psy-chological number,” or threshold, to give legitimacyand credibility to the sample results. Given the absenceof a list of polling stations, in order to reduce the arbi-trary choice of a polling station within a location, theobservers were given the instruction to select thefourth polling station from wherever they entered thegrounds of the polling location.

A coordinator, a technical assistant, a database man-ager, and 11 data entry operators staffed the centraloffice. Based on the sample and taking into account anequitable contribution of the participating organiza-tions, 11 provincial supervisors supported by 50provincial-level coordinators were selected, and theprovinces were subdivided in 38 zones (based on the dis-tribution of the sample points), each to be managed byone or two coordinators. The supervisors and coordina-tors were responsible for the identification and trainingof nearly 1,600 observers, preparation of a deploymentplan, and the practical implementation of the PVT.

The recruitment and training of the coordinatorsand observers were completed without major prob-lems, and detailed information was compiled on eachobserver (typically teachers, journalists, public servants,students, NGO staff, and church members) to mini-mize recruitment of “ghost” observers (e.g. observerswho exist in name only to collect pay but do not actu-ally do the work). The training sessions were designedby Maputo staff but implemented locally with randommonitoring. The Electoral Institute of Southern Africaassisted in the update of the observation and tabula-tion forms and in the development of trainingmaterials for the supervisors and coordinators.

Initially, the Electoral Observatory resisted theidea of observers calling Maputo directly before deliver-ing the results to their provincial coordinator,preferring that observers deliver the results to theircoordinator, who would then call Maputo. The CarterCenter was concerned that this latter approach would

FIGURE 4: GEOGRAPHIC

DISTRIBUTION OF PVTSAMPLE POINTS

Maputo city (47)Maputo province (47)Gaza (62)Inhambane (50)Sofala (64)Manica (44)Tete (56)Zambézia (148)Nampula (142)Niassa (45)Cabo Delgado (87)

___________________

Total (792)

THE CARTER CENTER


60

jeopardize the speed of results transmission as hap-pened in 2003. It might also result in more data beingcollected than necessary, demanding more sorting inthe Maputo central office. The eventual communica-tion system was based on the availability of telephoneand cellular phone service in the deployment areas ofthe observers. Observers would call Maputo directlyfrom their locations where network coverage was avail-able. Otherwise, where such networks were notavailable, the observers were to travel to the nearestplace with coverage and make their call to Maputo.

The Electoral Observatory did not develop a pub-lic information strategy to handle the PVT results.Several weeks before the election, the ElectoralObservatory convened a press conference to present itsplans, but there was no further public communicationuntil the Electoral Observatory leader, Brazão Mazula,issued a preliminary statement on Dec. 6. A detailedPVT report was delivered to the Mozambican electoralbodies, the main political parties, and presidential can-didates on Dec. 9, but due to scheduling difficulties,Renamo leader Afonso Dhlakama did not meet withthe Electoral Observatory until Dec. 17.

The observation was carried out without majordifficulties. With only a few exceptions, allobservers were present at their indicatedpolling location before opening of thepolling stations and remained throughoutthe vote counts. Approximately 100 of the1,600 trained observers could not bedeployed because of problems with theiraccreditation (mainly in Nampula, Niassa,and Zambezia). Despite the logistical difficul-ties that some observers encountered, datawere collected for 785 polling stations forthe presidential elections and 769 for the leg-islative elections, totaling, respectively, 99percent and 97 percent of the sample. Inwelcome contrast to 2003, the communica-tion plan was very well-implemented and,during the night after the closing of thepolls, the central office received results from

204 sample points. These results indicated a cleartrend (fluctuation of the result graph for the presiden-tial elections stabilized after about 150 data entries)but was urban biased, as the first results came exclu-sively from areas with mobile phone coverage. A morebalanced and reliable PVT result was reached within72 hours after the close of polls.

The provincial supervisors were responsible for col-lecting all report forms and bringing them personally tothe PVT central office in Maputo. In a sharp improve-ment from 2003, 70 percent of the forms reached thePVT office within four days of the elections.

PVT RESULTSThe PVT results proved to be consistent with the

official results, with a margin of error of less than 0.5percent. For the presidential elections, the PVT pro-jected 63.3 percent for Armando Guebuza and 31.8percent for Afonso Dlhakama, compared to the offi-cial results of 63.6 percent for Guebuza and 31.7percent for Dhlakama. For the legislative elections,the PVT reported 61.5 percent for Frelimo and 30percent for Renamo, while the official results gaveFrelimo 62 percent and Renamo 29.7 percent.

TABLE 5: COMPARISON OF OFFICIAL CNERESULTS AND PVT

RESULTS

CNE PVT

ArmandoGuebuza

63.6% 63.3%PRESIDENT

AfonsoDhlakama

31.7% 31.8%

Frelimo 62.0% 61.5%NATIONAL

ASSEMBLY Renamo 29.7% 30.0%

THE CARTER CENTER


61

The slightly wider margin of error for the legisla-tive elections, though acceptable, occurred because thesample was primarily based on a national scale for thepresidential elections and not on a provincial scale aswould have been formally required. The quality andreliability of the results were based on two factors: 1) athoroughly prepared and tested sample and 2) correctimplementation of the observation and rapid transmis-sion of results by the observers.

Data analysis in Maputo also revealed some seriousanomalies in the electoral process, such as proof of bal-lot stuffing in the district of Changara in Teteprovince. At a later stage, when the qualitative observa-tion forms were reviewed, additional findingsincluded:

■ Evidence of a very strong, proximate, and, often,armed police presence in practically all polling sta-tions, in contravention of the legal prescription thatpolice should not be closer than 300 meters of thepolling station. The domestic observers viewed this asa form of potential intimidation because the policeforce is so strongly identified with Frelimo.

■ All over the country, polling staff rejected voterswith a valid voter card if their name was not on thevoter register, contrary to official instructions that suchindividuals should be allowed to vote.

■ The collected data confirmed serious inconsist-encies in the official voter register and the implicationsfor the calculation of a correct rate of voter participa-tion.

Several institutional lessons were learned from theCarter Center’s partnership with the ElectoralObservatory, including:

■ Collaboration among participant organizationswas challenging. Collaboration can be experienced as athreat to individual resource mobilization, compound-ed by individual sensitivities and a working politicalenvironment where a lack of trust is common.

■ At times, management decisions produced ten-sions between the “political” leadership of the PVTand the technical staff.

■ Key provincial staff often accepted job offersfrom other organizations that jeopardized their avail-ability and, at times, even their required neutrality asPVT observers.

■ Participant organizations recruited observers onthe basis of organizational loyalty rather than objective-ly determined requirements (e.g. education level, beingregarded as a respected and neutral person, and livingin the area of deployment). This did not influence thePVT implementation negatively, but it did have costimplications as more observers required transport toget to their respective polling stations.

■ Difficulties with the transfer of money toobservers and problems with accreditation resulted inlast-minute reshuffling of the deployment plan.

■ The lack of experience with such complex andsensitive operations, especially at provincial level, put aheavy burden on the small central team to keep activi-ties focused and implemented within strict time lines.

The Electoral Observatory is commended for thisextraordinary achievement. The lessons from 2003 weretaken into account, absorbed, and corrected by thenational staff, in particular by a more thorough train-ing process and the maintenance of key staff in thecentral team. There is now national capacity to conducta PVT, but it depends very much on the involvementof less than a handful of technical staff who implement-ed two consecutive PVTs. This achievement now facesthe serious challenge of sustainability. One possibleapproach would be to maintain the ElectoralObservatory as the platform where member organiza-tions can discuss electoral and democracy issues, receivefurther training, and organize future activities.

CARTER CENTER SUPPORTThe Carter Center was pleased to play a support-

ing role to the leadership of national Mozambicanorganizations in this exercise, and many of the movingparts of the Center’s activities in Mozambique–fieldoffice, consultants, long- and short-term observers,and delegation leadership–interacted with theElectoral Observatory.

THE CARTER CENTER


62

The Center’s field representatives, Marc deTollenaere and Nicolás Bravo, worked closely with theElectoral Observatory on a range of election issues,including the PVT, deployment plans, and reporting.Election specialists Eric Bjornlund and Glenn Cowanprovided essential technical expertise to the project atcritical moments.

The Center’s long-term observers also workedclosely with domestic observers, exchanging informa-tion and contacts. During deployment of the Center’sshort-term delegation, each Center team met withdomestic observers prior to and, when possible, afterthe elections. The Center’s observers assisted in thedelivery of PVT observation forms to many PVT coor-dinators in central locations throughout Mozambique.In coordination with other international observers,mostly but not only from the European Union, theCenter collected results from 15 percent of the PVTsample points to serve as a cross-check.

The delegation leadership also met with BrazãoMazula and the other leaders of participant organizationsin the Electoral Observatory. The speed of the PVTimplementation enabled Center staff to brief PresidentCarter that Frelimo was headed to a substantial victoryand the observed irregularities were insufficient to over-turn the overall results. Given the tight electoral race

between Frelimo and Renamo in 1999, lingering uncer-tainties about the 1999 results process, and thesubsequent political deadlock in party relations, the PVTprovided an impartial verification of the official tabula-tion. The availability of the PVT results soon after theclose of polls allowed The Carter Center and otherobservers to make more informed judgments about theoverall picture.

The dedication of the Electoral Observatory mem-bers, the Center’s technical assistance, and the stronglinks forged between the two organizations generated amore unified political assessment of the 2004 elec-tions. The Center hopes that future activities willconsolidate the role of domestic observers and theskills acquired during the implementation of the firstnational PVT in Mozambique. Mozambican civil socie-ty actors can now share their experience with oneanother and their counterparts elsewhere in southernAfrica and further abroad. The Center hopes it helpedto create awareness of the importance of a PVT as anobservation tool and contributed useful technicalcapacity. Given the past conflict and current competi-tive political context in Mozambique and thesignificant problems of CNE and STAE election man-agement (e.g. politicized decision-making, lack of acredible voter register, and a delayed and confused tab-ulation process), the Center believes the successfulconduct of the PVT provided a technically reliable andindependent confirmation of the election results. ThePVT, therefore, served as a convincing conflict preven-tion tool and a crucial instrument for the promotionof how citizens can ensure the transparent conduct ofall aspects of the electoral process.

The Center believes the successful conduct of the PVT provided a technically reliable andindependent confirmation of the election results.

THE CARTER CENTER


63

On Dec. 1-2, 2004, Armando Guebuza waselected president of Mozambique, usheringin another five years of Frelimo-led govern-

ment. The Carter Center commends the individualsand institutions responsible for the 2004 elections,including members of the CNE, STAE, the govern-ment of Mozambique, and political parties, forensuring a peaceful election. Despite this achievement,the Center concludes that serious irregularities under-mined the achievement of a transparent and fairelection in all parts of Mozambique. The Centerbelieves that the definitive margin between the twomain presidential candidates and the confirmation ofofficial results by the independent parallel vote tabula-tion guarantee President Guebuza’s victory. But giventhe significant foreign investment in Mozambique’speace process and democratic development, and espe-cially in light of the ongoing concerns regarding thestate of the voter register and the conduct of tabula-tion, it is imperative that Mozambique implementmeasures to ensure procedurally correct and politicallyconvincing elections.

To these ends, the Center makes the followingsummary observations and recommendations in a spir-it of cooperation with the people and official electoralbodies of Mozambique to sustain their democracy.

1. Improve voter confidence. The CNE shouldthoroughly investigate irregularities, implement appropri-ate responses, and communicate the results to thepublic. Incidents of ballot box stuffing, lost electionmaterials, exclusion of votes in tabulation, tally sheetswith alterations or errors, and other problems under-mine public confidence in the election authority and thepolling process even when they do not affect the overallresults. The CNE’s opportunity to dispel remainingdoubts about the accuracy of official results is dimin-ished when observers are unable to view and understandthe reasons for rejected polling station tally sheets.

2. Improve credibility and accuracy of voterregister. For the many reasons outlined in this report,the voter register requires further review and consolida-tion in order to produce the clean and credible listMozambicans deserve. In addition, election authoritiesshould be able to manage this process in accordancewith the law on an annual basis and produce a listwith the number of registered voters per polling sta-tion. This list should be made available to politicalparties and others prior to election day. In February2005, STAE recommended not to update the voter reg-ister in 2005 but to take the time to clean the existingregister. This is a welcome suggestion that should beimplemented.

3. Reform CNE structure. The Carter Centerhas previously recommended review of the size andpartisan structure of the CNE. The appointment of acivil society representative as president of the CNE is awelcome move; yet it is insufficient to ensure that elec-toral supervision is conducted in a manner that isnonpartisan, transparent, and of service to allMozambicans regardless of party affiliation. A range ofoptions to minimize the infusion of party politics intothe CNE could be considered. The success of theConstitutional Council in appointing members whorenounce their party affiliation during their appoint-ment might be one approach to consider. At 19members, the CNE is expensive and too large for effec-tive decision-making.

4. Strengthen the multiparty system. In contrastto the close election between Frelimo and Renamo-UEin 1999, the 2003 and 2004 elections produced a sig-nificant consolidation of Frelimo’s strength andpopular support. There are many factors that explainthis success and the difficulties faced by smaller andnewer political parties. Yet a strong multiparty systemis a central component to sustainable democracy.Democratic leadership that embraces multiparty

CONCLUSIONS AND RECOMMENDATIONS

THE CARTER CENTER


64

cooperation and broad participation by civil societycould provide a foundation for improved governance.On some level, the current political culture ofMozambique discourages opportunities for effectivecollaboration. When the police display favoritism dur-ing the election campaign, when Renamo delegates areunable to remain with the ballot boxes overnight, orwhen Renamo representatives deliberately delay tabula-tion or other crucial election processes, not only theelectoral process but also the party system is under-mined. Greater professionalism and respect fordifferences of ideology or policy can mitigate the tenseelectoral environment yet allow for political debate.The code of conduct for political parties must be sup-ported not only at signing in May but on each day asthe election approaches.

5. Build transparent campaign finance. Politicalparties should comply with the currently establishedlaw for disclosure and transparency in political finance.Moreover, the potential for abuse of public resourcesand inadequate regulation of private sources of fundsshould be re-examined.

6. Strengthen positive strides in electoral dis-pute resolution. The Center welcomed the revisionof the electoral law to provide copies of polling stationtally sheets to party representatives. In addition to theposting of tally sheets outside polling stations, thisprovision allows parties to check official results.However, the fact that official results are announcedonly by district and not by polling station underminesthe effectiveness of the provision to resolve disputes.The establishment of the Constitutional Council andits mandate to review the final official results also pro-vided an appropriate legal recourse for electoralcomplaints. The Center is concerned that althoughthe council correctly rejected certain election peti-tions, it did so while acknowledging that the citedirregularities could impact the distribution of seats inthe National Assembly. Lingering questions about thecredibility of the results may negatively affect the oper-ation of Mozambique’s sovereign lawmaking body.The Center regrets that Renamo leader AfonsoDhlakama and his party’s 90 members of Parliamentboycotted Armando Guebuza’s inauguration as presi-dent on Jan. 28 yet they still chose to take their seatsin Parliament.

7. Increase voter participation. Mozambiquemust investigate the reasons for low voter turnout.STAE should update the voter register in a mannerthat indicates the true number of eligible voters andallow for the establishment of enough polling stationsto enable more voters to reach polling stations, espe-cially in rural areas. Election authorities, politicalparties, and others need to examine how they canencourage greater political participation among citizensand build public confidence in the effectiveness of theelectoral process.

Future reforms must ensure credible elections.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


65

8. Support participation of women and youth.The Center was encouraged by the significant presenceof women and youth as election officials and partyagents and hopes this participation continues toreceive the support it deserves. Frelimo also is congrat-ulated for its commitment to ensuring that womenand youth are well-represented on the party lists forthe National Assembly.

9. Hold one-day elections. The majority of pollingstation officials discharged their duties effectively andwithout prejudice. A two-day election is especially oner-ous on all participants, and Mozambique shouldexplore the means to hold single-day polls.

10. Increase transparency and effectiveness oftabulation. The tabulation of votes at provincial andnational levels was seriously hindered by Mozambique’sweak communications and transport infrastructure.Aside from the government’s general responsibility toimprove these public resources, election authorities canmitigate their effects through more careful planningand distribution of resources. Additional trainingemphasis on the counting process for election officialsand party agents also might ensure more consistentdetermination of valid and invalid ballots at thepolling-station level, reducing the workload at higherlevels. In addition to designing effective tabulationprocesses that respect the integrity of voter intent, elec-tion authorities should reassess how they value the roleof election observers. Arrangements that respect theintegrity of the results process while providing for maxi-mum transparency of tabulation and determination ofvalid polling station results can be designed. In theevent that Mozambique continues along the path of acomputerized voter register and results process, thesetechnology applications will only be effective if they areaccompanied by careful advance planning, good publiccommunication about their purpose, and well-trainedoperators implementing a process that enjoys the confi-dence of political parties, candidates, and voters.

11. Ensure timely announcement of results.None of the provincial election commissions nor theCNE met their respective legal deadlines during the

tabulation process. As the Constitutional Council hasinstructed, the CNE needs to develop a feasible sched-ule so provincial and national tabulation can finish ina timely manner. The Center encourages the CNE toannounce and publish complete polling station results.

12. Provide impartial election security. Reportsof police bias in favor of Frelimo during the electionperiod underscore the need to respect the right ofMozambicans to vote in an environment that is secure,tolerant, and impartial. Further training on the appro-priate presence of security on election day should beimplemented.

13. Build professional and impartial media. TheCenter’s observers received complaints about biasedmedia coverage, particularly regarding TVM, Noticias,and, notably, Domingos. Although allotted broadcasttime was proportionally available to all presidentialcandidates and parties, partiality in media coverageundermines the right of Mozambicans to be informedabout their political choices. While political parties areultimately responsible for communicating their mes-sages effectively, the public media can be especiallysupportive in this regard. Of note: Radio Mozambique,which has a national coverage and broadcasts in locallanguages, proved to be less partial and more profes-sional. Media workers and owners should receiveadditional training on aspects of the electoral processand law as well as how best to cover election campaignswithout political bias.

14. Support domestic observers. Past experiencein Mozambique and elsewhere has demonstrated thesignificant contribution of domestic observers to theintegrity of elections. The Center commends all non-partisan domestic election observers in Mozambiqueand is especially appreciative of the important collec-tive effort of the Electoral Observatory and the parallelvote tabulation. The PVT added to the credibility ofthe official results, and this experience should be builtupon for future elections.

15. Support international observers. The pres-ence of nonpartisan international observers indicatesan international interest–not desire to interfere–in

THE CARTER CENTER


66

Mozambique and should be fostered. The Centerunderstood that its invitation to observe the electionsincluded the campaigns, polling, and results process.Unfortunately, despite multiple assurances of access tothe activities and record of the tabulation process, TheCarter Center received very limited or inadequateinformation. Many other aspects of the Center’s inter-action with the CNE and STAE were positive, andThe Carter Center hopes to continue to work withMozambicans in a spirit of friendship and democraticsolidarity.

16. Sustain the Constitutional Council. In thebrief time since its establishment, the ConstitutionalCouncil has executed an important oversight role. Thecouncil should intensify its commitment to the timelyreview of complaints and work to ensure that the sub-mission process is clear to concerned political parties.While the submission procedure is important, it also iscrucial that the concerns of political parties bereviewed on substantive grounds and not be rejectedout of hand because of procedural errors.

Mozambique’s new president, Armando Guebuza, facesthe dual challenges of social and economic development.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


67

THE CARTER CENTER


68

2004 MOZAMBIQUE ELECTIONSHORT-TERM DEPLOYMENT

NameProvince ofDeployment

Route

1 Silvina Silva–Aras & Samuel Kivuitu Zambézia Quelimane to Mocuba

2 Zulmira Rodrigues & Bradley Austin Gaza Xai – Xai to Chókwe

3 Abdoulaye Kourouma & Harry Vanden Sofala Beira to Muanza

4 Achille Nisengwe & Nina Frankel Niassa Cuamba to Metarica

5 Natasha Cassinath & Torben Frandsen Inhambane Inhambane to Morrumbene

6 Jean-Paul Murekezi & Frances Henderson Cabo Delgado Mundimbe to Mueda

7 Gemima Neves & Mooroogessen Veerasamy Inhambane Vilankulo & surroundings

8 Fernanda Lopes, William Kleh, & Jane Nandy Cabo Delgado Pemba to Montepuez

9 Carrie Manning, Henry & Rebecca Tinsley Sofala Meringue to Caia

10 Maria Macchiaverna & Paul Nsapu Cabo Delgado Cabo Delgado North

11 Mário Jaleco & Amanda Dixon Tete Chifunde & surroundings

12 Jason Calder & Maria Helena Alves Nampula Ilha de Moçamique & Mossuril

13 John Fleming & Mario de Paiva Zambézia Mocuba to Gurue

14 Margot Gould & Cipriano Gomes Manica Chimoio to Sussendenga

15 Scott Taylor & Elma Doeleman Nampula Angoche & surroundings

16 Roel Borren & Wellington Chibebe Niassa Lichinga to Mandimba

17 Anna Bjorndal & Anne Pitcher Tete Changara & surroundings

18 Cecilia A. Luna López, Frances Johnson Morris,& Kathleen Hawthorne

Maputo Matola to Muamba

19 Patrick Berg & Ana Ganho Tete Tete & Moatize

20 Jacques Saidi–Kamuleta, James (Chip) Carter, &Rebecca (Becky) Carter

Nampula Nampula to Nacala

21 Girum Tesfaye & Carter Center Staff Maputo Maputo

Leadership1

President and Mrs. Carter, Dr. David Pottie,Nicolás Bravo, & Nancy Konigsmark

Maputo Maputo

Leadership2

President Nicéphore Soglo, Ambassador Mongbe,Mr. Glago, Marc De Tollenaere, & Dr. JohnHardman

Maputo Maputo

THE CARTER CENTER


69

Leadership, Teams 18, 21

Team 1

Teams 8, 6

Team 16

Team 3

Teams 11, 17, 19

Team 2

Team 5

Team 7

Team 14 Team 15

Team 20

Team 13

Team 10

Team 12

Team 9

Team 4

THE CARTER CENTER


70

POLL OPENING OBSERVATION FORMMozambique, December 1-2, 2004

Observer team: __________________

Date: ______________

A. INFORMATION ABOUT THE POLLING STATION

Province: District: Location: Polling stationnumber:

Arrival time:Departure time:

Time openingprocedure began:

Number of voters inline, if any:

Time first ordinaryperson voted:

B. OUTSIDE THE POLLING STATION (mesa) YES NO1. Is the area within 300m of mesa free of party posters/campaign activity?

2. Are security personnel at least 300m away from mesa?

C. POLLING STATION ATTENDANCE YES NO3. Are three or more mesa officials present? President ___ Vice-

Pres./Secretary____ Clerk 1 ____ Clerk 2 ____ Clerk 3 ___

4. Are there any women mesa officials? If yes, specify roles:

5. Do at least two mesa officials speak a local language?If yes, specify:

6. Are party agents from more than one party present?If yes, specify: Frelimo ___ Renamo- UE ____ PDD _____ other ___

7. Are domestic observers present?If yes, specify: FECIV ____ AMODE _____ other (list) _____

8. Are other international observers present?If yes, specify:

D. OPENING ON DECEMBER 1 YES NO

9. Were ballot boxes shown to be empty prior to being sealed?

10. Were boxes correctly sealed?

11. Were party agents able to record the seal numbers?

11. Do the electoral supply kits have only one number?

12. Are all required materials available? If no, specify: voter register ___ballots ___ ballots boxes ___ indelible ink ____ election forms ___ other___

13. Did the poll open on time at 7:00 am? If no, specify:

THE CARTER CENTER


71

D. OPENING ON DECEMBER 1 YES NO

14. Are party agents and/or observers able to observe process adequately?

15. Are mesa officials responsive to party agents concerns?

16. Was the opening free of disruption/restriction? If no, specify:

17. Was the opening free of formal complaints to the mesa officials? If no, specify:

18. Did mesa and security officials vote before opening of the poll?

E. OPENING ON DECEMBER 2 YES NO

19. Did the poll close on time last night? If no, specify time:

20. Was the last person in queue at 6:00 pm allowed to vote?

21. Were boxes sealed correctly?

22. Were seals undamaged at opening?

23. Did police guard ballot boxes at night?

24. Did party agents and/or domestic observers stay with the boxes overnight?

If yes, specify:

25. Are all required materials available? If no, specify: voter register ___

ballots ___ ballots boxes ___ indelible ink ____ election forms ___ other___

26. Did the poll open on time at 7:00 am?

If no, specify:

27. Did at least two party observers witness the reopening of the polling station?


29. Are mesa officials responsive to party agents’ concerns?

30. Was the opening free of disruption/restriction? If no, specify:

31. Was the opening free of formal complaints to the mesa officials? If no, specify:

F. ADDITIONAL COMMENTS OR PROBLEMS REPORTED Use the space below for any additional TCC team comments or problems reported to you and who reported it.

THE CARTER CENTER


72

POLLING OBSERVATION FORMMozambique, December 1-2, 2004

Observer Team: _________________

Date: ________________

A. INFORMATION ABOUT THE POLLING STATIONProvince: District: Location: Polling station

number:

Arrival time:Departure time:

Repeat visit?YES / NO

Time poll opened: If Dec 2, time openedyesterday:

Voters on register(max. 1000):

Number alreadyvoted:

Number in lineto vote (if any):

Total number ofballots received:

B. OUTSIDE THE POLLING STATION (mesa) YES NO1. Is the area within 300m of mesa free of party posters/campaign activity?

2. Are security personnel at least 300m away from mesa?

C. POLLING STATION ATTENDANCE YES NO3. Are three or more mesa officials present? President ___ Vice-

Pres./Secretary____ Clerk 1 ____ Clerk 2 ____ Clerk 3 ___

4. Are there any women mesa officials? If yes, specify roles:

5. Do at least two mesa officials speak a local language?If yes, specify:

6. Are party agents from more than one party present?If yes, specify: Frelimo ___ Renamo- UE ____ PDD _____ other ___

7. Are domestic observers present?If yes, specify: FECIV ____ AMODE _____ other (list) _____

8. Are other international observers present?If yes, specify:

D. POLLING OPERATION YES NO9. Does the mesa President explain the ballot procedure to each voter before

handing him or her a ballot?

10. Are all ballot boxes sealed and numbered?

11. Are all required materials available? If no, specify: voter register ___ballots ___ ballots boxes ___ indelible ink ____ election forms ___ other___

12. Are voters’ fingers checked for signs of ink as they enter the mesa?

THE CARTER CENTER


73

D. POLLING OPERATION YES NO

13. Are voter cards checked against the voter register?

14. Are voters able to keep their vote secret during the entire voting process?

15. Does the mesa clerk sign the voter register next to each voter’s name after they

cast their ballot?

16. Are voter cards returned to the voter before leaving the mesa?

17. Are voters’ fingers inked before leaving the mesa?

18. Does the mesa President sign and keep all spoiled ballots?

19. Do mesa officials appear to be adequately trained and knowledgeable about

their role?


21. Are mesa officials responsive to party agents’ concerns (if any)?

22. Was the mesa free of disruption/restriction of the voting process? If no, specify:

23. Was the poll free of formal complaints to the mesa officials? If no, specify:

E. ON DECEMBER 2, if possible, inquire about; YES NO

24. Did the poll close on time last night? If no, specify time:

25. Was the last person in queue at 6:00 pm allowed to vote?



28. Did police guard ballot boxes at night?

29. Did party agents and/or domestic observers stay with the boxes overnight?

If yes, specify:

30. Did at least two party observers witness the reopening of the polling station?

F. OVERALL EVALUATION Carter Center Mesa

officials

Domestic

observers

Party

agents

31. Mesa functioned well, no problems

32. Minor problems, unlikely to impact on result

33. Significant problems, potential for impact on

result

34. Serious violations, should invalidate results

G. ADDITIONAL COMMENTS OR PROBLEMS REPORTED Use the space below for any additional TCC team comments or problems reported to you and who reported it.

THE CARTER CENTER


74

POLL CLOSING, COUNTING AND RESULTSOBSERVATION FORM

Mozambique, December 1-2, 2004

Observer team: ____________ Arrival time: _______ Departure time: ________

A. INFORMATION ABOUT THE POLLING STATIONProvince: District: Location: Polling station

number:

B. CLOSING ON DECEMBER 1 YES NO

1. Did the poll close on time? If no, specify time:

2. Was last person in queue at 6:00 pm allowed to vote (if applicable)?


4. Were police prepared to guard ballot boxes overnight?

5. Were party agents and/or observers prepared to stay overnight with the ballotboxes?

C. CLOSING ON DECEMBER 2 YES NO


7. Did the poll close on time at 6:00 pm?

8. Were all voters in line at 6:00 pm allowed to vote (if applicable)?

C. COUNTING ON DECEMBER 2 YES NO

9. Is there sufficient light for counting?

10. Are spoiled and unused ballots checked and packed in their respective bags?

11. Is the number of actual voters checked on register?

12. Are ballot boxes opened and the number of ballot papers inside counted (bothboxes)?

13. Does Secretary note the number of ballot papers in the edital?

14. Does President display and read out loud each ballot?

15. Are ballot papers arranged by candidate/party ? Blank and void ballots shouldalso be counted and arranged separately.

16. Does Secretary record the number of votes cast for candidate/party?

17. Does President confirm the total votes cast reconciles with the number of actualvoters as shown on the voter register?

18. Does President invite party agents to examine the stacks of ballots?

20. Are editais filled in correctly, signed and stamped by the polling station membersand posted outside the polling station?

21. Do party agents receive copy of the edital?

22. Do party agents sign the acta?

23. Was the counting free of disruption?

24. Was there a full discussion in each case of a disputed ballot (if applicable)?

THE CARTER CENTER


75

25. Did all participants agree with all decisions?

D. PRESIDENTIAL RESULTS

Results directly observed ____ or recorded from posted editais______ (check one)

TOTAL

Armando Guebuza (Frelimo)

Afonso Dhlakama (Renamo)

Raul Domingos (PDD)

Yagub Sibindy (PIMO)

Carlos Reis (UNAMO)

Valid (válido) ballots

Invalid (nulo) ballots

Blank (branco) ballots

Spoiled (inutilizado) ballots

Disputed ballots (reclamado and/or protestado)

E. LEGISLATIVE RESULTS

Results directly observed ____or recorded from posted editais______ (check one)

TOTAL

Frelimo

Renamo

PDD

List parties as applicable

Valid (válido) ballots

Invalid (nulo) ballots

Blank (branco) ballots

Spoiled (inutilizado) ballots

Disputed ballots (reclamado and/or protestado)

F. ADDITIONAL COMMENTS OR PROBLEMS REPORTED Use the space below for any additional TCC

team comments or problems reported to you and who reported it.

THE CARTER CENTER


76

PRE-ELECTION STATEMENT ON MOZAMBIQUE ELECTIONS,JULY 26, 2004

FOR IMMEDIATE RELEASEMonday, July 26, 2004

CONTACT: Marc de TollenaereIn Mozambique, 258-082-31-18-81

MAPUTO, MOZAMBIQUE… In response to an invi-tation issued by the National Elections Commission(CNE), The Carter Center observed the voter registra-tion update in Mozambique from June 28 to July 15.With observers from five countries—Belgium, theDemocratic Republic of the Congo, Ecuador, Guinea-Conakry, and the United States—The Carter Centervisited 151 registration brigades in nine provinces.

The Center is awaiting the release of the final reg-istration update figures before issuing an overallassessment. However, the Center has several prelimi-nary observations.

VOTER REGISTRATION UPDATEThe mission found that registration officers were

generally well-trained and well-equipped to performtheir task and worked harmoniously with party moni-tors. Toward the end of the registration period, somebrigades ran out of materials when new supplies didnot arrive due to transport problems, but no caseswere observed where voters could not register at all.

Carter Center observers noted regional differencesin the participation rate of urban and rural voters.Even though each brigade is responsible for approxi-mately the same number of voters, observers found inthe southern provinces that urban brigades tended toregister nearly twice the number of voters per day asthose in rural areas. By contrast, the discrepancy was

higher in the central provinces with urban brigadesregistering voters at a daily rate four times that of thosein rural areas.

Although it is difficult to identify the exact causesfor these regional differences, observers found varyinglevels of civic education and long distances to registra-tion brigades may have been contributing factors.Voters often had to walk considerable distances, andofficials from the Technical Secretariat for ElectoralAdministration (STAE) regularly had too few vehiclesto transport materials or to supervise the operation ofthe brigades effectively. While the location of some ofthe nearly 2,500 brigades may have been an issue, thepolitical parties approved these through their represent-atives on the district, provincial, and national levels ofthe electoral authorities. The Center will continue tofollow these and other issues related to the voter regis-tration update.

COMPUTERIZED VOTERS ROLLThe Center was pleased to find the computerized

registration books of 1999 and 2003 were generally avail-able for verification of voter details at the registrationbrigades. However, only a limited number of votersappeared to review their inscription data. The Center isparticularly concerned about several problems regardingthe credibility of the computerized roll. Center observersnoted problems with the computerized registers in 10 to

THE CARTER CENTER


77

15 percent of the visited brigades. In some cases, thebooks appeared to have database errors such as repeti-tion of voter names, birth dates, or numbers, and insome cases voter names were not listed. The CarterCenter previously noted some of these problems follow-ing its observation of the 2003 municipal elections, butthese ongoing errors imply the subsequent correctionscarried out by STAE have yet to produce a fully cleanvoters list.

The Carter Center strongly encourages electoralauthorities to take all possible measures to continueimproving the voter register to avoid multiple voter reg-istration or disenfranchisement of eligible voters. Thededication of adequate resources, thorough cross-checking of brigade supervisor reports, and referenceto the manual voter registration books are amongmeasures to improve the credibility of the voters roll.Consideration also should be given to conducting anindependent audit of the voter register.

ROLE OF INTERNATIONAL OBSERVERSThe Carter Center would like to thank those local

electoral officials who greeted its observers with open-ness. However, the Center also urges the CNE andSTAE to better inform provincial, district, and techni-cal staff about the rights, duties, and role of election

observers. Although properly accredited, Centerobservers often were received with distrust and insecu-rity on the part of local electoral officials. Manydistrict and local officials were unaware also of CNEregulations on election observers. Greater understand-ing of the role of international observers could beestablished during the training of election officials andby ensuring the proper distribution of CNE observa-tion regulations to district and local officials.

The Carter Center will observe the ongoing com-pilation of the voters roll and encourages electoralauthorities to ensure all eligible Mozambicans have theopportunity to participate in the December election.

The Carter Center was founded in 1982 by formerU.S. President Jimmy Carter and his wife, Rosalynn, inpartnership with Emory University, to advance peaceand health worldwide. A not-for-profit, nongovernmen-tal organization, the Center has helped to improve lifefor people in more than 65 countries by resolving con-flicts; advancing democracy, human rights, andeconomic opportunity; preventing diseases; improvingmental health care; and teaching farmers to increasecrop production. Visit www.cartercenter.org to learnmore about The Carter Center.

THE CARTER CENTER


78

ATLANTA…. Former U.S. President Jimmy Carter,his wife, Rosalynn, and former Benin PresidentNicéphore Soglo will lead a 60-member internationaldelegation to observe Mozambique’s presidential andlegislative elections. The Carter Center, whichobserved the 1999 national elections and the 2003municipal elections, was invited by the NationalElection Commission and welcomed by all major polit-ical parties to observe the Dec. 1–2 elections.

The Center’s election observation mission followstwo assessment trips in 2003, during which electionauthorities, political parties, and local observer groupsalso welcomed the Center’s presence. The Center alsohas engaged in Mozambique through the Center’sGlobal Development Initiative, which has supportedthe national consensus-building initiative known asAgenda 2025.

“While Mozambique has introduced importantelectoral reforms since 1999, The Carter Center sharesthe concern of other observers about the transparencyof the tabulation process,” said Dr. David Pottie, sen-ior program associate of the Center’s DemocracyProgram. “International and regional standards haveestablished that election observers need access to allcritical phases of elections, including tabulation, inorder to credibly carry out their work. The Carter

Center hopes that appropriate conditions will be creat-ed in Mozambique so that observers are able to fullyassess the process."

President and Mrs. Carter, President Soglo, Dr.Pottie, and Nicolas F. Bravo, Mozambique field officerepresentative, hope to meet with presidential candi-dates, the election commission, domestic observers,and other international observers.

A team of nine long-term observers was deployedin early October to observe the political environment,election preparations, and the political party cam-paigns. The Center’s observers visited more than 50urban and rural districts in 11 constituencies. Theymet with representatives of the political parties, elec-toral authorities, domestic and international electionobservers, civic organizations, media, and internationalcommunity representatives.

Generally, observers found a calm environment,though some isolated signs of intimidation wereobserved in Gaza and Tete provinces. The code ofelectoral conduct signed by all political parties inMay 2004 was respected to varying degrees through-out the country, and the Center hopes it will bewidely adhered to during the last period of the electoral process.

Center observers found their exchanges with

CARTER CENTER TO OBSERVE MOZAMBICAN ELECTIONS

FOR IMMEDIATE RELEASETuesday, November 16, 2004

CONTACT: Kay TorranceIn Atlanta, 404-420-5129

Carter Center Field OfficeIn Mozambique, 258-1-497949

THE CARTER CENTER


79

international and domestic observers to be helpful andencourage them to continue their key efforts in the elec-toral process. The Carter Center also welcomes theefforts carried by Technical Secretariat for ElectoralAdministration (STAE) to consolidate the computerizedvoters roll and looks forward to its further progress.

The remainder of the delegation, representing 23countries, arrives Nov. 27 and will receive briefings inMaputo before deployment throughout Mozambique.On Dec. 1 and 2, Center observers will witness pollopenings, voting, poll closing, and counting and trackthe tabulation of results.

The Carter Center was founded in 1982 by formerU.S. President Jimmy Carter and his wife, Rosalynn, inpartnership with Emory University, to advance peaceand health worldwide. A not-for-profit, nongovern-mental organization, the Center has helped to improvelife for people in more than 65 countries by resolvingconflicts; advancing democracy, human rights, andeconomic opportunity; preventing diseases; improvingmental health care; and teaching farmers to increasecrop production. Please visit www.cartercenter.org tolearn more about The Carter Center.

THE CARTER CENTER


80

The Carter Center was invited by the NationalElection Commission (CNE) and welcomed by allmajor political parties to observe the Dec. 1–2 elec-tions. The Center observed the 1999 and 2003elections and has been engaged in initiatives inMozambique, including support for the Agenda 2025national consensus-building initiative and agricultureproduction technologies through SG 2000.

Under the leadership of former U.S. PresidentJimmy Carter, his wife, Rosalynn, and formerPresident of Benin Nicéphore Soglo, the Centerdeployed 60 international observers from 23 countriesto 11 provinces for the elections. The observers metwith local officials, campaign teams, and domesticobservers and observed the voting, counting, and ini-tial tabulation. The delegation leadership met inMaputo with President Joaquim Chissano and all pres-idential candidates except Carlos Reis, as well asmembers of the CNE, Technical Secretariat forElectoral Administration, and the ConstitutionalCouncil; leaders of nonpartisan domestic electionmonitoring organizations; and others. We would liketo extend our thanks to all of the many other individu-als and organizations who welcomed our observationefforts and took the time to facilitate our understand-ing of Mozambique’s politics and electoral process.

ASSESSMENT OF THE ELECTIONThe Carter Center congratulates the people and

leaders of Mozambique for the conduct of the presi-dential and legislative elections. To date, with a fewexceptions, our assessment of Mozambique’s electionsis positive. We are especially pleased about the peacefulatmosphere that prevailed on election day and thecalm and orderly manner in which the poll was con-ducted in most places.

This election marks another important step inMozambique’s ongoing democratization. Because thetabulation and verification of final results are ongoing,it is too early to evaluate the election as a whole. TheCenter will continue to observe these processes in thedays and weeks ahead and will maintain its long-termmonitoring through the announcement of final results.After the conclusion of the entire electoral process, theCenter will issue a more comprehensive report.

THE POLLING PROCESSCarter Center observers visited nearly 1,000

polling stations in all 10 provinces and Maputo. Ourobservers generally found the polling stations theyvisited were well-organized, functioned effectively,were fully staffed, and had necessary election materi-als. We were impressed especially with the thoroughand consistent application of the polling procedure

FOR IMMEDIATE RELEASESaturday, Dec. 4, 2004

CONTACT: Nicolás BravoIn Maputo, cell 082-308-004

POSTELECTION STATEMENT ON MOZAMBIQUE ELECTIONS,DEC. 4, 2004

THE CARTER CENTER


81

and the significant number of women polling offi-cials. However, we noted the distribution of pollingstations in some districts resulted in some voters hav-ing to travel long distances to get to the polls.

Previously, The Carter Center raised concernsabout the accuracy of the voter register, with manyobsolete names still listed. Although no systematicproblems regarding the voter register were apparent onelection day, the number of registered voters perpolling station should have been made available topolitical parties and others prior to election day.

The opening and closing process was well-managed.Our observers found the counting procedures were cor-rectly applied with meticulous attention to detail.However, there seemed to be some unnecessary voidingof ballots, and additional training might ensure moreconsistent determination of valid and invalid ballots.Transparency in the counting process has beenstrengthened by the welcome introduction of providingcopies of the final tally sheets to party agents.

Preliminary results indicate low voter turnout.Although it is too early to explain this worrying trend,the Center hopes that election authorities, politicalparties, and others will encourage greater political par-ticipation and build public confidence in theeffectiveness of the electoral process.

ELECTION OBSERVERSAND CANDIDATE WITNESSES

The Carter Center urges electoral authorities tofoster transparency and credibility of results by provid-ing nonpartisan domestic and international observersfull access to ongoing tabulation of results and con-firming of reasons for any invalidation of votes cast.

Past experience in Mozambique and elsewhere hasdemonstrated the significant contribution that domes-tic observers and effective candidate witnesses canmake to the credibility and integrity of the electionprocess. We commend the important work and com-mitment of the member organizations participating inthe Electoral Observatory.

In light of the controversies about the final elec-

tion results in 1999, domestic observers have correctlyplaced additional emphasis on counting and consoli-dating the results in this election. The quick countconducted by the Electoral Observatory will provide anindependent check of tabulation and results, with asmall margin of possible error.

Carter Center observers were generally well-received by election officials, political party representa-tives, and observers. We sought to conduct ourobservations without interference in the normal con-duct of the polls and found presiding officers werewilling to answer our questions regarding the process.Unfortunately, several Carter Center observer teamswere made to feel unwelcome by the Tete provincialelection commission. The presence of internationalobservers from many organizations, including theEuropean Union, Commonwealth, Southern AfricaDevelopment Community Parliamentary Forum,Electoral Institute of Southern Africa, and others, indi-cates an international interest in Mozambique thatshould be fostered.

Our observers reported that more than one partyor candidate agent was present at most polling sta-tions. Party agents have a direct interest in ensuringthe integrity of the polling process, and their effortsshould continue to receive support. In some districts,however, Frelimo tended to be the only party effec-tively represented at the polls. Elsewhere, someopposition Renamo party agents had difficulty exer-cising their right to remain with the ballot boxesovernight on Dec. 1.

POLICE AND SECURITYThe Center is concerned about several incidents

that resulted in the arrest of Renamo party agents andsupporters in Angoche. We also received reports ofpolice bias in favor of Frelimo during the election cam-paign and on the election days. In some cases, anexcessive number of ballots were voided when the willof the voters seemed obvious. The Carter Centerhopes future political activity in Mozambique willenjoy more secure, tolerant, and impartial enforcement

THE CARTER CENTER


82

of the law with the guidance of an effectively appliedcode of conduct for political parties.

CONCLUSIONThe electoral process is not complete, and the

Center will continue to observe the ongoing tabula-tion of results at provincial and national levels. TheCenter encourages election officials to ensure everyeffort will be made to enable transparency in the offi-cial results process. It is important that there arejustified reasons for the invalidation of any ballots ortally sheets. The establishment of the ConstitutionalCouncil and its mandate to review the final officialresults provides an important reassurance that the con-cerns of all participants in the elections will beaddressed.

The Carter Center trusts that the operation of theCNE, Constitutional Council, and Mozambique’sother electoral institutions will ensure the choice ofthe Mozambican people is reflected in the final results.

The Carter Center was founded in 1982 by formerU.S. President Jimmy Carter and his wife, Rosalynn, inpartnership with Emory University, to advance peaceand health worldwide. A not-for-profit, nongovern-mental organization, the Center has helped to improvelife for people in more than 65 countries by resolvingconflicts; advancing democracy, human rights, andeconomic opportunity; preventing diseases; improvingmental health care; and teaching farmers to increasecrop production. Visit www.cartercenter.org to learnmore about The Carter Center.

Carter Center field representative Nicolás Bravo and President Carter confer with an election official before the opening of polls.

GR

AN

TL

EE

NE

UB

ER

G

THE CARTER CENTER


83

Continuing its observation of the Dec. 1-2 presidentialand legislative elections, The Carter Center has moni-tored the tabulation of results in provincial capitalsand the reclassification of invalid votes at the NationalElections Commission (CNE) in Maputo.

The Center has followed several issues since theelection days that were of concern, including the arrestof Renamo representatives in the Manica, Niassa, andNampula provinces; delayed poll openings in ruralareas of Zambezia; and low voter turnout. The Centerwill maintain its presence in Mozambique until theconclusion of the elections and then will publish acomprehensive report.

The Carter Center is very concerned about a num-ber of irregularities observed during the provincialtabulation, including polling station tally sheets resultslacking credibility, problems with the tabulation soft-ware, mismatched numbers of polling stations andtally sheets, and mistrust between political party repre-sentatives in the provincial Technical Secretariat forElection Administration (STAE). No province submit-ted results by the legal deadline, and the Center foundit difficult to acquire clear and reliable information onthe progress of tabulation. These and other concernsprovide evidence of serious weaknesses inMozambique’s vote tabulation, and as a result, the

Center cannot verify the accuracy of the provincialcounting process at this time.

The CNE reviews rejected tally sheets sent toMaputo from the provinces in a closed plenary session.Former U.S. President and Carter Center Chair JimmyCarter has previously asked CNE to maintain a clearrecord of reasons for the rejection or acceptance ofthese tally sheets and to allow nonpartisan observers toexamine the rejected tallies. The Center repeats itsrequest for full observer access to this record to con-tribute to the transparency and credibility of the finalresults process.

The Center was impressed by the generally peace-ful postelection environment and the genuinecommitment of most Mozambicans to the legalrequirements of vote tabulation.

Although the Center does not expect these irregu-larities to alter the overall outcome of the presidentialelection, they do undermine the credibility ofMozambique’s electoral authorities. In addition, anyerrors in the tabulation of legislative election resultscould lead to the incorrect distribution of seats. TheCenter is confident the findings of the parallel votetabulation carried out by the domestic observer groupsin the Electoral Observatory serve as a useful registeragainst which to assess the overall election results.

FOR IMMEDIATE RELEASETuesday, Dec. 21, 2004

CONTACT: Maputo Field Office+ 258-1-497949

POSTELECTION STATEMENT ON MOZAMBIQUE ELECTIONS,DEC. 20, 2004

THE CARTER CENTER


84

PROVINCIAL TABULATIONCarter Center observers generally were granted

access to review the entry of polling station results atcomputer terminals in the provincial electoral commis-sions, although this access was uneven, and in oneprovince, police and electoral officials obstructed theCenter’s observer. In another province, only one com-puter terminal was available to observers, and districtresults were not available for review. Also of impor-tance, observers were unable to examine rejected tallysheets in all provinces.

Tabulation was delayed in all provinces, and noprovince met the legal requirement of submission ofresults to the CNE by Dec. 9. A variety of reasons wereadvanced for the delays, including late reception ofmaterial from districts, tabulation software flaws, andmisunderstandings between political party representa-tives. Provincial counting has been characterized bypolitical mistrust, and in some provinces, Renamo rep-resentatives to the STAE obstructed the opening ofwarehouses with election materials.

In a number of cases, Carter Center observersfound tally sheets with unrealistically high voterturnout, including multiple instances of polling sta-tions in Niassa and Tete recording a 100 percentturnout and more than 90 percent support forFrelimo. Given the low turnout nationwide (rangingfrom 30-40 percent), ballot boxes appear to be stuffedat polling stations in the Tete districts of Changara,Chifunde, and Tsangano as well as in the Niassa dis-tricts of Metarica and Marrupa and in the Gazadistrict of Chicualacuala. In some of these areas,Carter Center observers had reported an intimidatingenvironment during the campaigns, and oppositionparty agents had problems in getting credentials.

A final polling station list with registration booknumbers and numbers of registered voters was nevermade available to political parties or observers. Thislist, described by the CNE as a “state secret,” is essen-tial because it determines the number and location ofpolling stations and should coincide with the database

used for the tabulation software. The reasons for suchsecrecy are not clear, and it is still unknown how manyregistered voters or physical polling stations existed onelection day. Errors in the software resulted in the gen-eration of extra tally sheets, creating confusion forpolitical parties and election observers. According toSTAE, the “phantom tally sheets” were produced whendata were entered from polling stations with morethan one voter register book. It appears that STAE hastaken steps to address this problem, but its existenceconstitutes a fertile ground for allegations of fraud.Following the 1999 and 2003 elections, The CarterCenter recommended timely production of credibleand secure tabulation software.

As a result of these and other difficulties in receiv-ing clear and full information, the Center is unable toverify the accuracy of the provincial counting processat this time.

NATIONAL TABULATIONThe Center’s observers had adequate access to the

reclassification of invalid ballots at CNE headquarters,although on some occasions it was limited to specifictimes during the day. All invalid ballots are sent to theCNE in Maputo for reclassification where CNE repre-sentatives from Frelimo and Renamo, working inteams of two, examine each ballot paper. The Centernoted a striking incidence of invalid ballots fromNiassa and Tete provinces with a consistently appliedpattern of additional ink marks that were mostly seenon ballots that would have otherwise been forRenamo. Necessarily, both party representatives pres-ent at the reclassification room considered thoseballots void.

The Center is aware that Renamo’s leadership hasannounced it will not accept the results. Althoughfinal results were due Dec. 17, the Center encouragesall candidates and their supporters to continue to par-ticipate in a climate of dialogue and to use legalchannels for the resolution of any election disputesthat may arise following the CNE’s announcement.The Center also welcomes the positive involvement of

THE CARTER CENTER


85

nonpartisan Mozambican civil society organizations,such as members of the Electoral Observatory, andencourages their assessment of the tabulation andresults process. As noted in previous statements, theCenter anticipates the Constitutional Council willensure that valid concerns with the conduct of theelections will be addressed properly.

The Carter Center was founded in 1982 by formerU.S. President Jimmy Carter and his wife, Rosalynn, inpartnership with Emory University, to advance peaceand health worldwide. A not-for-profit, nongovernmen-tal organization, the Center has helped to improve lifefor people in more than 65 countries by resolving con-flicts; advancing democracy, human rights, andeconomic opportunity; preventing diseases; improvingmental health care; and teaching farmers to increasecrop production. Please visit www.cartercenter.org tolearn more about The Carter Center.

A party agent signs and receives a copy of the polling station tally sheet at completion of count.

SIG

RID

SPIN

NO

X

THE CARTER CENTER


86

The Carter Center recognizes the overall results andcongratulates the elected Frelimo President ArmandoEmílio Guebuza. However, the Center concludes theNational Elections Commission (CNE) has not admin-istered a fair and transparent election in all parts ofMozambique. Political parties must also be heldaccountable since it is their representatives in the CNEand the Technical Secretariat for ElectionAdministration (STAE) who are responsible for theoverall success or failure of the elections. The Centerhas attempted to observe and assess as much of the ver-ification process as possible but has been hindered by alack of cooperation by the CNE.

The Center’s previous statements following theelection reported positive and peaceful election daysbut expressed concerns over the accuracy of the votersregister, some irregularities on the polling and the tab-ulation process, and the fact that the list withregistration book numbers of registered voters wasnever available to political parties or observers.

The Carter Center welcomes the ConstitutionalCouncil’s announcement Jan. 19 validating the finalresults of the Dec. 1-2 elections. The council’sannouncement underlines the need to create adequateconditions for electoral observation in Mozambique.It also stresses the poor level of professionalismdemonstrated by electoral authorities and the politicalparty representatives who assumed roles in the elec-toral institutions.

The overall election results are not in question, asindicated by the wide margin of Frelimo’s victory andconfirmed by the parallel vote tabulation conducted bydomestic observers. However, the problems observedby The Carter Center could have had serious conse-quences in a closer election. Moreover, the Centerremains concerned that the Constitutional Council hasvalidated CNE election results retaining irregularitiesthat could have had an impact on the distribution ofparliamentarian seats in some provinces. Despite for-mer U.S. President Jimmy Carter’s Dec. 3 request toCNE President Arão Litsure that observers reviewrejected tally sheets and a detailed record of the reasonsfor their rejection, the Center and other observers havenot been able to view such a record of the rejection of699 presidential and 731 legislative tally sheets. Unlessand until the CNE provides clear evidence to dispel anyremaining doubts about the accuracy of official results,the Center believes the credibility of the tabulationprocess will remain open to question.

A comprehensive election report, including recom-mendations for electoral reform, is forthcoming. Thisis the Carter Center’s fourth and final public state-ment on the Mozambique 2004 elections.

BACKGROUNDIn October 2003 the Center opened a Maputo

office and monitored the November 2003 municipalelections. In partnership with the member organiza-

FOR IMMEDIATE RELEASEWednesday, Jan. 26, 2005

CONTACT: Nicolás BravoIn Maputo, cell 082-308-004

POSTELECTION STATEMENT OF THE CARTER CENTER

ON THE MOZAMBIQUE ELECTIONS, JAN. 26, 2005

THE CARTER CENTER


87

tions of the Electoral Observatory, the Center provid-ed technical assistance to conduct a parallel votetabulation (PVT) in 10 of Mozambique’s 33 municipal-ities. The Center’s public statements on the municipalelections voiced concerns regarding the accuracy of thevoter register and the need for more transparent elec-tion administration. The Center also called attentionto the uneven observer access to the provincial votetabulation and the reclassification of invalid votes con-ducted by the CNE in a closed session. Over thecourse of 2004, the Center’s staff in Maputo remainedactive, strengthening links with Mozambican civil soci-ety organizations, political parties, media, and theelectoral bodies while organizing activities in supportof the electoral process.

The Center also observed the voter registration inJune-July 2004, again voicing concerns regarding thecredibility of the computerized voter roll, difficulties inobtaining observer access to critical phases of theprocess, and regional discrepancies in the registrationrate in favor of urban voters over rural voters. At thetime, the Center suggested an independent audit ofthe voter register could help to improve the credibilityof the Mozambican electoral bodies.

In early October 2004, the Center deployed ninelong-term observers to observe the entire process,including the civic education and political campaigns.While the political environment was generally peace-ful, some intimidation was observed in certain districtsin the Tete, Gaza, and Niassa provinces. The Centerwas particularly concerned about some police agentswho demonstrated partiality and unequal treatmenttoward opposition party supporters. A climate of secre-cy within the CNE, marked by persistent failure torelease a definitive list of polling stations indicating thenumber of registered voters, also was encountered inthe provinces, raising questions about the CNE’s com-mitment to observer access.

During the election days, the Carter Center obser-vation mission, headed by former U.S. PresidentJimmy Carter, his wife Rosalynn, and former Presidentof Benin Nicéphore Soglo, deployed 60 observers from

23 countries throughout the entire country. The dele-gation leadership met in Maputo with PresidentChissano and key actors of the electoral process.

While the overall assessment of the election dayswas positive, the Center made clear in a Dec. 4 state-ment that the tabulation process was unfinished and afinal assessment would be released at a later date.From the nearly 1,000 polling stations visited,observers noted that they were generally well organ-ized. One major incident observed by the Centerresulted in the arrest of Renamo party agents and sup-porters in Angoche, while a similar situation wasobserved in Quelimane. The Center’s observers in Tetefound that at least one domestic observer carrying outhis PVT duties was arrested in the district of Zumboand remained unreachable for several days. In 90 per-cent of the visited polling stations, the Center notedthe presence of at least two party agents, though someof the polling stations in Gaza and Tete provinces wereobserved with only one party agent.

Following three weeks of monitoring postelectionprocesses, the Center released a second interim state-ment Dec. 21, emphasizing its continuing concernsabout the number of irregularities observed during theprovincial tabulation, including questionable tallysheet results, serious problems with the software, mis-matched numbers of polling stations and tally sheets,and a constant mistrust between political party repre-sentatives at the STAE. The newly introducedprovision for party and candidate agents to sign andreceive copies of polling station tally sheets was a wel-come step toward a more accountable results process.However, these copies appear not to have been used asa cross-check against rejected tally sheets at later stagesof the tabulation. No Provincial ElectoralCommission released clear information on the finalvoter list, and software problems resulted in the gener-ation of extra tally sheets, creating confusionexacerbated by vague responses from the electoralauthorities. While the explanation for such errorsgiven later by the STAE seemed plausible, their exis-tence created fertile ground for allegations of fraud.

THE CARTER CENTER


88

Evidence of serious irregularities in the pollingprocess came to light in several provinces. For example,the Center’s observers in the Tete provincial districtsof Changara, Chifunde, and Tsangano as well as in theNiassa districts of Metarica and Marrupa and in theGaza district of Chicualacuala found voter turnout per-centages suspiciously high and even in some cases,impossibly high (more than 100 percent), leading tothe conclusion that ballot stuffing occurred in some ofthose polling stations. These incidents had an impor-tant impact in the final National Assembly results.Also notable, the province of Tete had the highest (andunprecedented) voter turnout nationwide (67.4 per-cent), contrasting with a national rate of 43.6 percent.

While noting the multiparty environment provid-ed an important opportunity for all parties to worktogether to improve governance, the Center urged theCNE to take steps to ensure the prompt and transpar-ent verification of results.

TABULATION OF FINAL RESULTSThe Center has serious unanswered questions

about the complete accuracy of the results and the lackof transparency in the CNE’s final tabulation. Forexample, the results did not include a detailed district-by-district map, and the CNE has poorly explained thereasons for rejected, stolen, or missing tally sheets.Despite assurances to observers that they would begranted full access to a detailed record listing the rea-sons for rejected polling station tally sheets, thisinformation has not been made available.

The Center has attempted to observe and assess asmuch of the verification process as possible but hasbeen hindered by a lack of cooperation by the CNE. InNovember 2004, President Carter requested observeraccess to national tabulation but was told the CNE ple-nary sessions are closed. President Carter subsequentlyrequested that observers be present for the CNE reviewof invalid ballots sent from polling stations to Maputo.Although the CNE initially granted this request,observer access was subsequently restricted to two hoursa day. Carter Center observers were able to follow the

data entry of computerized results, but the software didnot enable searches by polling location. Despite theCNE’s assurance to President Carter that observerswould be allowed to review a record of the reasonsgiven for rejected tally sheets, the CNE has not pro-duced this record, and the Center has been unable toexamine even a sample of rejected tally sheets.

The Center also is alarmed by the apparent lack ofinterest in the tabulation of results displayed by non-represented political parties and domestic observers.Given the concerns about the credibility of previouselection results, the Center expected these groupsmight have been more assertive about the right to atransparently conducted exercise.

The Center’s confidence in the overall electionresult is based in large part on the successful conductof Mozambique’s first national PVT. The Center con-tinued its assistance to the Electoral Observatorythroughout 2004 and is pleased the PVT closelymatched the official results. Mozambique’s domesticelection observers are to be congratulated for the con-duct of this crucial check on the official results.

PETITIONSFour days after the legal deadline, the CNE

announced the official election results Dec. 21. Theelectoral law requires election petitions to be filed withthe CNE within two days of announcement of resultsand any appeals to be filed with the ConstitutionalCouncil within five days of the CNE’s decision.

The main opposition parties, Party for Peace,Democracy and Development (PDD), IndependentParty of Mozambique (PIMO), the EnlargedOpposition Front (FAO), the Movement for Changeand Good Governance (MBG), and the Renamo-Electoral Union (Renamo-UE), prepared formalelection petitions. PDD wrongly presented its petitionfirst to the Constitutional Council and was thereforerejected. PIMO, MGB, and FAO called for the annul-ment of the elections, but their petitions also wererejected by the CNE.

On Dec. 27, Renamo-UE presented a petition to

THE CARTER CENTER


89

the CNE calling for new elections within six months.The CNE sent the petition to the ConstitutionalCouncil without deliberation. The council consideredthis action inappropriate and sent the petition back tothe CNE, demanding formal deliberation. The CNEfinally rejected the Renamo complaint Jan. 3, 2005,which called for new elections within a period of sixmonths. The CNE correctly argued the petition wassubmitted after the legal deadline.

Following the CNE deliberation and ruling, theConstitutional Council rejected the Renamo-UE com-plaint on Jan. 15. The council noted the Renamo-UEcomplaint was submitted after the legal deadline, thealleged irregularities should have been reported at thetime and location they occurred, and the petition’scontent, as presented to the council, was in fact a newpetition since it differed from the one originally pre-sented to the CNE. On Jan. 12, the ConstitutionalCouncil also rejected complaints of the minor parties.

FINAL RESULTSAlthough the CNE announced overall results Dec.

21, they were only made publicly available to media andobservers five days later. The CNE’s final report (acta)was short on details explaining the reasons and originsof rejected tally sheets, and no district results wererecorded as required by law. Nonetheless, in DeliberationNo. 19 of Jan. 19, 2005, the Constitutional Council vali-dated the results in this form.

The Carter Center welcomes the strong recom-mendations made by the Constitutional Council,including the generation of a single, national voter reg-ister, the proper institutional and professionaldevelopment of the CNE, a more consistent knowl-edge of the electoral legislation on the part of thepolitical parties, and the need to create adequate con-ditions for electoral observation.

Nevertheless, the Center remains concerned thatsome issues did not receive sufficient attention from thecouncil, including the abuse of public resources by polit-ical parties during the campaigns, acceptance of thedelayed results from the electoral bodies but no such

flexibility being accorded to political parties with lateelection petitions, and the reluctance of some district-level electoral bodies to receive party complaints(especially since the council cited Renamo-UE’s failureto submit electoral complaints at the district level as oneof the three arguments for rejecting its petition). Thecouncil registered its concerns that despite evidence ofconsiderable irregularities, a culture of impunity pre-vails, and those responsible are not held accountable.

Although the council notes the CNE failed to com-plete a final tabulation map of district-by-district resultsfor the provinces of Nampula, Manica, Sofala, andGaza, in clear violation of the electoral law, the CNE isnot held accountable for its failure to provide detailedreasons for the rejected polling station tally sheets.

CONCLUSIONSMozambique is at a critical point in its democratic

development. While it is clear the people ofMozambique have endorsed Frelimo and its presiden-tial candidate, Armando Emilio Guebuza, the Centeris concerned the enduring problems with the voter reg-ister, evidence of serious irregularities and fraudduring polling in several provinces, and inadequatetransparency of the tabulation process will continue tocast a shadow over Mozambique’s democracy.

Democratic leadership that embraces multipartycooperation and broad participation by civil societycould provide a foundation for improved governancein Mozambique. The Center hopes Mozambique’spolitical leaders will take steps to meet this goal andurges them to respond with renewed commitment toelectoral reform.

The Center makes these observations with noauthority and no intention of intervening inMozambique’s affairs but in the spirit of supportingdemocratic development in Mozambique and else-where. Ultimately, it is the Mozambican people whowill judge the legitimacy of the election and will holdgovernment and officials accountable.

THE CARTER CENTER


90

Overview: The Carter Center was founded in 1982 byformer U.S. President Jimmy Carter and his wife,Rosalynn, in partnership with Emory University, toadvance peace and health worldwide. A nongovern-mental organization, the Center has helped to improvelife for people in more than 65 countries by resolvingconflicts; advancing democracy, human rights, andeconomic opportunity; preventing diseases; improvingmental health care; and teaching farmers to increasecrop production.

Accomplishments: The Center has observed 53 elec-tions in 24 countries; helped farmers double or triplegrain production in 15 African countries; mediated orworked to prevent civil and international conflictsworldwide; intervened to prevent unnecessary diseasesin Latin America and Africa; and strived to diminishthe stigma against mental illnesses.

THE CARTER CENTER AT A GLANCE

Budget: $38 million 2003-2004 operating budget.

Donations: The Center is a 501(c)(3) charitable organi-zation, financed by private donations from individuals,foundations, corporations, and international develop-ment assistance agencies. Contributions by U.S. citizensand companies are tax-deductible as allowed by law.

Facilities: The nondenominational Cecil B. DayChapel and other facilities are available for weddings,corporate retreats and meetings, and other specialevents. For information, 404-420-5112.

Location: In a 35-acre park, about 1.5 miles east of down-town Atlanta. The Jimmy Carter Library and Museum,which adjoins the Center, is owned and operated by theNational Archives and Records Administration and isopen to the public. 404-865-7101.

Staff: 150 employees, based primarily in Atlanta.

MA

RT

INF

RA

NK

ONE COPENHILL

453 FREEDOM PARKWAY

ATLANTA, GA 30307

404-420-5100 ◆ FAX 404-420-5145

WWW.CARTERCENTER.ORG

OB

SE

RV

IN

GT

HE

2004

MO

ZA

MB

IQ

UE

EL

EC

TIO

NS

Documents

Observing the 2004 Mozambique Elections - The Carter · PDF fileObserving the 2004 Mozambique Elections Waging Peace. Fighting Disease. ... São Sebastião Cahora Bassa Dam Licoma