Electoral Fraud and its Preconditions

1
Electoral Fraud and its Preconditions

• Kiril Kolev
• Graduate Student Prospectus Workshop Presentation
• 10/19/2008

2
Overview

• Electoral Fraud Definition
• Measuring Electoral Fraud
• Existing Theoretical Arguments
• Existing Empirical Work
• The Project Model and Hypotheses
• Data
• Fieldwork

3
Definition of Electoral Fraud

• Electoral fraud is an excessive political
  pressure in the form of intentional manipulation
  of the electoral process that is in violation of
  the law. It could occur both in the pre-election
  and election period.

4
Measuring Electoral Fraud

• There are several sources of information that
  could be used to learn about the extent of fraud
• Public opinion
• Political Parties
• Media reports
• Independent monitoring organizations

5
What precipitates Fraud?Existing theoretical
arguments

• Concentrated economic interests ability and
  willingness of businesses to deliver votes for
  parties that they find favorable to their
  interests
• Socio-economic Population density, income levels
  and equality, urban vs. rural
• Political system/party characteristics single
  vs. multi-member districts, PR vs. FPTP,
  presidential vs. parliamentary, internal party
  discipline, party control over electoral lists
• Dominant party stability If a dominant
  authoritarian party perceives threat to its
  power, it is more likely to engage in fraud

6
Existing Empirical Work

• Case studies Focus on a single country and
  explore the instances of fraud and speculate
  about their causes.
• Large-n studies Simpser (2004, 2005) studies the
  effect of fraud on turnout and margin of victory
  in both semi-democratic and non-democratic
  regimes. Main problem is that he never employs
  fraud as a DV

7
The Model DVs

• General assessment by IEMs This is an indicator
  of whether the election represented the will of
  the people and to what extent there were
  problems.
• Specific cheating in the pre-election period
  Improper use of public funds Restricting the
  freedom of opposition parties to campaign
  Restrictions on media Pre-election intimidation
  Overall level of pre-election cheating
• Specific cheating on election day Vote
  processing Voter Fraud Intimidation Overall
  level of election-day cheating

8
Hypothesis 1

• H1 Countries that have experienced substantial
  electoral fraud in the past are more prone to
  having flawed elections in the future.
• IVs of interest include
• Presence/level of fraud in previous elections

9
Hypothesis 2

• H2 Political system characteristics, such as
  electoral laws, ballot format, and party
  control/discipline will influence the propensity
  for having a flawed election. Majoritarian,
  personalistic systems that exhibit low party
  cohesion and control over candidates will be more
  prone to electoral fraud.
• IVs of interest include
• FPTP/SMD vs. PR/multimember systems
• Regime type pure presidentialism vs.
  separation-of-powers
• Voter list composition closed- vs. open-list
• Internal party organization, including
  intraparty competition, party discipline/cohesion,
  etc.

10
Hypothesis 3

• H3 Young /unstable democracies will be more
  likely to experience fraudulent elections.
• IVs of interest include
• First multiparty elections
• Time since first multiparty election
• Regime stability

11
Hypothesis 4

• H4 The presence of international pressure from
  democratic countries or organizations, coupled
  with membership or aid conditionality will reduce
  the likelihood of fraudulent elections.
• IVs of interest include
• Application for membership in an international
  regime/organization at the time of election
• Membership in an international
  regime/organization at the time of election
• Receipt of foreign aid at the time of election

12
Hypothesis 5

• H5 Certain demographic and socio-economic
  characteristics of the population will influence
  the probability of fraud both across and within
  countries.
• IVs of interest include
• Socio-economic variables, such as income per
  capita, income inequality
• Sociological, such as monolingual share of
  population, effective number of religious and
  ethnic groups
• General, such as population density, balance of
  urban vs. rural share of the population.

13
Data

• Source Dataset on International Election
  Monitoring (DIEM), by Judith Kelley
• Coverage 1,169 national-level elections in
  1975-2004, 484 monitored by one of sixteen
  organizations, 106 by more than one
• Information coded All general, pre-election, and
  election-day DVs used above

14
Fieldwork

• Purpose Explore within-country relationships
  between the variable specified in the models
• South Africa and variation on the variables
  discussed above
• DV (Fraud) Elections have been classified as
  acceptable with major problems (1994), free and
  fair with minor problems (1999), free and fair
  with no problems (2004)
• IV 1 (electoral system) Mixed-member with half
  of 284 municipalities chosen through PR, half
  through FPTP
• Great variation in terms of socio-economic and
  demographic variables
• Recent transition to democracy

15
The Model IVs

• Independent variables
• Political Variables Electoral Institutions,
  Structures and Practices
• Electoral laws SMD vs multimember PR vs FPTP
  presidential vs. parliamentary
• Internal party organization and discipline
  party cohesion nominating process campaign
  financing number of same-party candidates
  competing for a seat
• Socio-Economic
• Concentrated economic interests A measure of
  labor/business strength. A possible proxy here
  could be union density/degree of unionization
• General sociological distribution of urban vs.
  rural, monolingual share of population, effective
  number of religions and ethnic groups, population
  density
• Socio-Economic controls Income per capita, GDP
  per capita, trade openness, foreign investment,
  income inequality (regional and national),
  education, civic organization density, etc.
• Regime-specific
• Government stability
• Democratic history Time since democratic
  transition should be included.
• Level of regime democracy POLITY or some
  related measure
• International influence Countries that are more
  exposed to international actors, in particular
  those who come from democratic countries,
• Conditionality This variable will code whether
  a country is applying for membership or aid from
  regional governance (e.g. EU) or economic (i.e.
  IMF) institutions.
• Monitoring Some countries will have a longer
  history of allowing international monitors of
  elections. This variable will code whether a
  particular country has had a low, medium, or high
  level of international monitoring going on while
  conducting elections. This could be measured as
  days spent by monitors per election.