Monday, 10 August 2015

Mexico faces decline in electoral integrity

By Miguel Angel Lara Otaola
(Visiting fellow Electoral Integrity Project, University of Sydney)

Historically Mexico has long experienced controversies over problems of electoral integrity, including malpractices of vote-buying and intimidation (Magaloni 2006, 2010; McCann, J. A., and Jorge I. Dominguez, 1998). The most serious issues, which, amongst other factors, guaranteed the PRI’s predominance, were thought to have been corrected during recent decades, although at the same time problems of narco-crime’s involvement in politics have resurfaced and the 2012 Presidential elections saw some street protests over allegations of voter corruption.

In the light of this history, on January 2014, Mexico’s Congress approved far reaching political and electoral reforms. This involved many constitutional changes and a new Electoral Code. Amongst other things, it allowed having coalition governments, increased checks on the President, expanded voting rights for citizens living abroad, and included special provisions for achieving gender parity in Congress. Moreover, it introduced the participation of independent candidates and allowed the re-election of law makers

The electoral reform enacted substantial changes. First, the Federal Electoral Institute (IFE) was transformed into a National Electoral Institute (INE). This new institution is in charge of organising federal elections (for the Presidency and Congress) and local elections (in coordination with 32 existing state election commissions). As a national body, it is now in charge of overseeing political party finance at the national level and of appointing the members of the country’s local election commissions. Furthermore, the reform changed the structure of INE’s Council (by adding two Councillors), strengthened its auditing powers, and increased sanctions for violations to the election code, including more provisions for the annulment of an election. In short, all these represented a significant evolution of Mexico’s electoral system.

The main stated aim of this change was to improve election integrity, especially confidence in electoral processes. In particular, as argued by its makers, the change from IFE to INE would bring local elections up to the internationally recognised standards of federal elections. Lorenzo Córdova, President of INE said that the reform sought to strengthen elections in all states

Yet it is unclear whether the reform will achieve its goals. In particular, data from expert Perceptions of Electoral Integrity (PEI) shows that in the 2012 Presidential election the country’s integrity score was 62 out of 100 points. By contrast, after the changes were implemented, in the 2015 legislative election, the score was 52. This is a modest but statistically significant drop in expert evaluations (R. .33, N. 34, P=.056). Moreover, this was not simply the difference between the types of contest, since legislative elections are usually evaluated slightly more highly by PEI experts than executive contests. 

In terms of relative global position, out of 153 elections evaluated between 2012 and 2015, this meant that Mexico dropped from position 55 to 95. This score is obtained from experts that monitor the quality of an election based on 49 indicators covering 11 components of the electoral cycle (from election laws and voter registration to election results). 

At the country level, Mexico ranks 54th of 125 and obtains a score of 57, which places it in the middle of the Latin American and Caribbean region (See Figure 1). Outside the region, countries with similar scores are India, Serbia, Ghana, Indonesia and Hungary. States such as Tunisia, Mongolia, Lesotho and South Africa -with their substantial challenges in electoral infrastructure- fare better than Mexico. 

Specifically, if we deal with the eleven components of the index, all but one (election results) have slightly decreased for Mexico. Out of these, however, only the drop in voting processes is statistically significant (P=.000) while the drops in voter registration, campaign media and vote count are significant but only at a 10% threshold. By contrast, evaluations of electoral results improved, largely due to the fact that the 2015 elections were perceived as being less challenged and having generated fewer protests. Figure 2 below shows the PEI score changes for all the parts of the electoral cycle. 

Figure 1. Electoral Integrity in Latin America and the Caribbean.

Figure 2. Changes in Electoral Integrity. Mexico, 2012-2015.

The overall drop must not be considered a direct effect of the recent political and electoral reform as many other factors may be at play. As several scholars have noted (Barreda and Ruiz Rodriguez, 2013; Birch, 2008), the general performance of the government can affect trust in other institutions, such as electoral processes. And these past two years have not been especially rosy for Mexico. To start with, on September 2014 a group of 43 people from the Ayotzinapa Teachers Training School were attacked and disappeared by municipal police, never to be heard from again. Soon after, a scandal erupted surrounding the buying of a multi-million-dollar house by the country’s first lady. Many other similar events have followed. And, if this was not enough, in July 2015 the top drug lord leader in the country Joaquin “El Chapo” Guzmán escaped (for the second time) from a high security prison

As a result, support for the political system has suffered greatly. According to the PEI experts, in the last three years confidence in the government, the parliament and the courts has dropped. Out of these, the drop in confidence in government is the only statistically significant (P=.054). While in 2012 confidence in the government was the highest amongst the three institutions (5.14 points on a 1 to 10 scale), in 2015 it was the lowest (3.8 points). Moreover, when tested further, confidence in government among Mexican experts were strongly and significantly related to Perceptions of Electoral Integrity, even after controlling for several other factors, such as the expert’s background (age, sex, education, and whether they are based in the country or international).

Elections are about technical accuracy – but also about broader perceptions and beliefs. Elections do not happen in a vacuum but they are embedded in a wider political system. A ´below average´ election organised in a country with a stable and legitimate political system has much more chances of being seen as credible than a technically perfect election in a country with a fragile –or momentarily weakened– political system. Let us hope that the storm Mexico is living passes swiftly. 


Barreda, M. and Ruiz Rodriguez, L. (2013) ‘La cadena causal de la confianza en los organismos electorales de América Latina: sus determinantes y su impacto sobre la calidad de la democracia’, Revista De Ciencia Política (Santiago), 33 (3), 649-673.

Birch, S. (2008). ‘Electoral institutions and popular confidence in electoral processes: A cross-national analysis’, Electoral studies, 27 (2), 305-320.

Magaloni, Beatriz. 2006. Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico. Cambridge: Cambridge University Press.

McCann, J. A., and Jorge I. Dominguez. 1998. ‘Mexicans react to electoral fraud and political corruption: An assessment of public opinion and voting behavior.’ Electoral Studies 17, (4): 483‐503

Norris, Pippa; Martinez i Coma, Ferran; Gromping, Max; Nai, Alessandro, 2015, "Perceptions of Electoral Integrity, Version 3.5",, Harvard Dataverse, V1

Schedler, Andreas. 2012. The Politics of Uncertainty Sustaining and Subverting Electoral Authoritarianism. CIDE: Mexico City.

Friday, 7 August 2015

The Year in Elections, mid-2015 update (PEI 3.5)

The latest version of the Perceptions of Electoral Integrity (PEI) Index has been released. This last version (3.5) includes the responses for 1696 experts for 153 elections in 125 countries. The data can be downloaded free of charge here. It is available at the expert, election and country level, and adds to the previous releases (3.0, 2.5, 2 and 1).

In this post, we present some interesting results for the 27 elections in 18 countries considered during the first half of this year. Before, though, we would like to thank all the experts that have devoted their time and knowledge answering our questionnaire. We define an expert as a political scientist (or social scientist in a related discipline) who has published on (or who has other demonstrated knowledge of) the electoral process in a particular country. Specifically, demonstrated knowledge is defined by the following criteria: (1) membership of a relevant research group, professional network, or organized section of such a group; (2) existing publications on electoral or other country-specific topics in books, academic journals, or conference papers; and (3) employment at a university or college as a teacher. At least forty experts per country were contacted for each election, including both domestic and international experts. The domestic/international distinction was made based on institutional affiliation, citizenship, and country of residence.

The elections included in this release took place between 1 January and 30 June 2015. They are the legislative elections of Uzbekistan, Greece, Comoros, Lesotho, Estonia, Tajikistan, El Salvador, Micronesia, Israel, Nigeria, Finland, Benin, United Kingdom, Guyana, Ethiopia, Suriname, Mexico, Turkey and Denmark and the presidential contests of Sri Lanka, Croatia, Zambia, Uzbekistan, Sudan, Togo, Kazakhstan and Poland. You can access the new executive summary EIP report here.

How was the integrity of those contests assessed by the PEI experts? We ask experts to evaluate elections using 49 indicators, grouped into eleven categories reflecting the whole electoral cycle. The dataset also includes a summary 100-point PEI Index based on summing all 49 indicators. These are the results for the overall PEI Index:

Finnish, Danish and Estonian elections perform very well in the ranking, well above the average (55.8). Perhaps, one of the most surprising results is the performance of the United Kingdom very similar to Benin’s. Another African country, Lesotho, performs above average.

There are two regions that seem to concentrate the worst performers of the elections: sub-Saharan Africa and former soviet republics. Indeed, Ethiopia, Togo and Sudan, jointly with Tajikistan and Uzbekistan perform very low in the index. Ethiopia’s performance, unfortunately, becomes the lowest of the series (disbanking 2013’s Equatorial Guinea). Obviously, democracy and development levels may have an impact on the performance but as the cases of Lesotho and Benin show, nothing is written in stone and the integrity of the elections can be improved.

Two big countries - Mexico and Turkey - perform below the average and worse than in previous years. Mexico loses about 10 points compared to the presidential elections of 2012, while Turkey looses about 4. It is also worth to mention that experts differentiate among the quality of elections within the same countries in the same year. A look to the evaluations of the Uzbeks Presidential and Legislative elections are a clear indication of this. While in neither any of the elections reaches the average, the Presidential (49) elections were substantively better evaluated than the legislative (39).

Obviously, this is just a first cut of the results, descriptive statistics without looking for any causal relation. In forthcoming posts we will show other results of the data. For the moment, you can download the data here and use it for your own research.

By the end of 2015, we will have gathered more than 175 elections and about 150. Keep posted for the new posts and updates. If you have any query, doubt or spot any inconsistency in the data, please do not hesitate to contact us at

Dr. Ferran Martinez i Coma
Sydney, 8 August 2015