When Guardians Matter Most: Exploring the Conditions Under Which Electoral Management Body Institutional Design Affects Election Integrity

Abstract

Problems with election fraud and election integrity are of increasing interest in both established and transitional democracies. In many transitional democracies, independent electoral management bodies (EMBs) have been championed as a key institutional reform measure to successfully strengthen election integrity. However, empirical findings regarding the impact of EMB institutional design on election integrity are mixed. While regional studies have found a positive impact of independent EMBs on election integrity in Latin America and Africa, global comparative studies appear to show that EMB institutional design is either negatively, or only very weakly related to election integrity. In this paper, we examine the effects of EMB institutional design on election integrity using the new Varieties of Democracy dataset and data from the International IDEA. We find that the mixed findings on EMB institutional design are due to the differences between transitional and established democracies on the one hand, and regimes with low and high quality of government on the other. The paper concludes with a reflection on results and a discussion of implications of these findings for the debate on electoral reform in Ireland.

Keywords:

• election integrity
• electoral malpractice
• election management body
• institutional design

Notes

1. Election integrity is a complex concept and as a consequence many different definitions abound. As such, conceptualisations vary from ‘positive’ definitions (emphasising the presence of desirable properties of elections) such as the free and fairness of elections, election quality or election integrity (Elklit & Svensson, 1997; Elklit & Reynolds, 2005; Lindberg, 2006; Norris, 2014, 2015) to ‘negative’ definitions (emphasising the absence of desirable properties) such as electoral manipulation, electoral malpractice, and election fraud (Schedler, 2002, 2013; Lehoucq, 2003; Birch, 2011;Simpser, 2013). Conceptualisations also differ in the normative criteria used to evaluate the quality of elections, ranging from international legal norms, to national law, to democratic theory (Lehoucq, 2003; Munck, 2009; Davis-Roberts & Carroll, 2010; Norris, 2014, 2015). See van Ham (2014) for an overview and discussion of the different conceptualisations and measurements used to study election integrity. Since our current analysis includes electoral regimes worldwide, ranging from electoral autocracies to liberal democracies, we use the term election integrity as it is based on international standards for elections. Election integrity refers to elections that ‘respect international standards and global norms governing the appropriate conduct of elections' (Norris, 2015: 4).

2. The Electoral Integrity Project collects data on 49 indicators of election integrity worldwide, based on country expert ratings. The data for Ireland and other countries can be found here: Perceptions of Electoral Integrity dataset: https://dataverse.harvard.edu/dataverse/PEI. For further information about the project, see: https://sites.google.com/site/electoralintegrityproject4/home.

3. IDEA distinguishes three models of electoral management: governmental, mixed and independent. Under the governmental model of electoral management, ‘elections are organised and managed by the executive branch through a ministry (such as the Ministry of the Interior) and/or through local authorities' (p. 7). The independent model of electoral management implies that ‘elections are organised and managed by an EMB that is institutionally independent and autonomous from the executive branch of government.' (p.7). This means that EMBs manage their own budget and are not accountable to the executive, though they are in practice often accountable to the legislature, the judiciary or the Head of State. Finally, in the mixed model of electoral management, the organisation of elections and implementation of electoral legislation will generally lie with a government department and/or local government, while policy, monitoring and supervisory functions are located in an organisation that is independent of the executive (p. 8). Note that this is a very rough classification of EMBs, and in practice many more differences in functions and accountability exist between different countries, as studies by IDEA (2006, 2014), Lopez-Pintor (2000) and Elklit & Reynolds (2001) illustrate. For the purpose of this paper however, we will follow this coarse distinction.

4. Of the 20 countries that changed the formal institutional structure of their EMB between 2006 and 2014, 16 changed towards increased formal EMB independence. This includes seven countries that went from a governmental EMB to an independent EMB (Egypt, Jordan, New Zealand, Sri Lanka, Syria, Tonga, and Tunisia); eight countries that changed from a mixed EMB to an independent EMB (Burundi, Cuba, Djibouti, Ivory Coast, Madagascar, Maldives, Togo and Cameroon); and one country that moved from a governmental EMB to a mixed EMB (Iran). Four countries changed in the opposite direction though: Saint Kitts and Nevis and Vanuatu changed from an independent EMB to a mixed EMB, Saint Vincent and the Grenadines changed from independent to governmental, and Monaco changed from mixed to governmental (IDEA 2006, 2014).

5. As Wall et al. (IDEA 2006: 11) note: ‘The performance of EMBs also depends on whether there is political will and commitment to allow the EMB to act freely and impartially, and on how the members of the EMB actually behave'.

6. Election integrity was measured using the V-Dem expert data, asking experts: ‘Taking all aspects of the pre-election period, election-day, and the post-election process into account, would you consider this national election to be free and fair?' Answer categories were the following:

0: No, not at all. The elections were fundamentally flawed and the official results had little if anything to do with the ‘will of the people' (i.e. who became president; or who won the legislative majority).

1: Not really. While the elections allowed for some competition, the irregularities in the end affected the outcome of the election (i.e. who became president; or who won the legislative majority).

2: Ambiguous. There was substantial competition and freedom of participation but there were also significant irregularities. It is hard to determine whether the irregularities affected the outcome or not (as defined above).

3: Yes, somewhat. There were deficiencies and some degree of fraud and irregularities but these did not in the end affect the outcome (as defined above).

4: Yes. There was some amount or human error and logistical restrictions but these were largely unintentional and without significant consequences.

Coppedge et al. (2015a). For more information about the project, codebook and data, see: https://v-dem.net.

7. Note that this pattern is strengthened if we consider election integrity for elections having taken place between 1990 and 2012: in such a larger sample, independent EMBs appear to score even lower in terms of average election integrity.

8. For more information about the project, codebook and data, see: https://v-dem.net.

9. EMB autonomy is measured by asking experts ‘Does the Election Management Body (EMB) have autonomy from government to apply election laws and administrative rules impartially in national elections?'.

10. Election integrity and malpractice has been measured in many different ways, ranging from election observation reports to media analyses, to ‘fraud forensics’ to perceptions by experts and citizens. In recent years several cross-national datasets have been developed that measure election integrity, mostly based on coding election reports, news media and historical sources (Elklit & Reynolds, 2005; Lindberg, 2006; Kelley & Kiril, 2010; Birch, 2011;Hyde & Marinov, 2012; Van Ham, 2012; Donno, 2013; Schedler, 2013; Simpser, 2013). Assessments of electoral malpractice based on ‘election forensics', as well as citizen surveys that measure perceptions of election integrity – are also increasingly common (Levin & Alvarez, 2012; Norris, 2013). Finally, most recently several datasets that collect expert perceptions of election integrity have been developed, including the Perceptions of Electoral Integrity data (Norris et al., 2014), and the Varieties of Democracy dataset (Coppedge et al., 2015b). While expert judgments can have validity problems, such as insufficient expertise or ideological bias, a recent study analysing potential sources of bias in expert evaluations of election integrity found very little evidence of systematic bias (Martinez i Coma and van Ham, 2015). Moreover, the V-Dem data used here are based on multiple expert ratings that have been aggregated using a Bayesian latent variable measurement model, in order to correct for potential coding error by experts (Pemstein et al., 2015). Finally, the correlation between the V-Dem measure of election integrity used here and the Freedom House electoral process indicator is 0.9, further increasing our confidence in the validity of our dependent variable.

11. EMB capacity is measured by asking experts ‘Does the Election Management Body (EMB) have sufficient staff and resources to administer a well-run national election?'. For more information, please see the V-Dem codebook (Coppedge et al., 2015a).

12. The replication data and do-file are available upon request from the authors.

13. Note that in model 2 and 3, EMB institutional design appears to have a significant effect on election integrity. In model 2, the coefficient for the main effect of EMB institutional design shows its effect in the base category for democracy: free regimes. Without control variables, it seems that in free regimes, the difference between independent EMBs and governmental EMBs is significant, and in the opposite direction: independent EMBs are associated with lower levels of election integrity. Yet once control variables are added in model 4 (and the combined model 6), it becomes clear that the differences between EMBs of different institutional design are not significant (if anything, even in free regimes independent EMBs are associated with slightly higher levels of election integrity than governmental EMBs). A similar pattern can be seen in the quality of government models presented in models 3 and 5 (and the combined model 6).