This article is part of the Democracy Futures series, a joint global initiative with the Sydney Democracy Network. The project aims to stimulate fresh thinking about the many challenges facing democracies in the 21st century.
The University of Sydney’s Electoral Integrity Project has just released its annual report, The Year in Elections 2015. Drawing on evidence from 54 countries that held elections in 2015, this report helps us understand their levels of compliance with international standards of electoral integrity. This is based in part on where and why governments attempt to restrict election monitors.
Since the 1980s, election monitoring has become widespread. It has become an international norm, with documents setting out the principles and code of conduct for both foreign and national election observers.
These codes call upon governments to guarantee access for such watchdogs. The reality is that monitors face restrictions in many places.
The 2015 Ethiopian election is a case in point. The African Union – which has a record of being rather uncritical of electoral manipulators – fielded the only international observers. The European Union and all other international groups were not invited. Civil society groups and independent media were harshly suppressed.
Similarly, human rights advocates were jailed and harassed in the run-up to Bangladesh’s 2013 election. In Russia, election watchdogs face legal action and an ever-shrinking public space for their advocacy. And Azerbaijan’s crackdowns on independent non-government organisations forced the Organisation for Security and Co-operation (OSCE) office to close, disabling its monitoring effort.
Evidently, there is a world of difference between the high hopes and dire realities of poll-watching.
“Top-down” factors may drive the supply of election monitors and their degree of access. If, for example, a government strongly wishes to be accepted by its regional peers, monitors may be less restricted in regions with overall stronger democratic linkage. Countries can use aid conditionalities, diplomacy, shaming and other international pressures to promote standards of electoral integrity.
“Pseudo-democrats” may have their own incentives to allow international observers or domestic watchdogs to operate. Most likely they want to signal that they adhere to international norms as one of the “good guys”.
However, “bottom-up” factors may also be important in explaining where monitors operate or are hindered. Democratic aspirations among a population may drive the desire for transparent elections, leading to calls for international attention and the creation of more watchdog NGOs. This could lead to a heightened engagement of international monitoring groups in a “boomerang effect”.
Public grievances are another factor. If elections work well, or if citizens believe parliamentary oversight or the courts can remedy any malpractice, the demand for monitors will be low.
However, where malpractices are rampant and accountability mechanisms are ineffective, grievances over unfair elections are likely to run deep. Here, one would expect popular demands for independent monitoring. Unfortunately, even where grievances exist, restrictions of rights of association and other civil liberties are certain to limit opportunities for watchdog NGOs and make their formation less likely.
What happens when observers are banned?
The Year in Elections 2015 report gives us some insight into these issues. It evaluates the integrity of all 180 national parliamentary and presidential contests held between July 1, 2012, and December 31, 2015, in 139 countries (including 54 in 2015).
The figure below plots two items of the survey of Perceptions of Electoral Integrity (PEI) used as the basis of the report. More than 2000 election experts were asked how far they agreed on a five-point scale with the statements “international election monitors were restricted” and “domestic election monitors were restricted”.
Countries are grouped by the overall integrity of their elections (low to very low; moderate; or high to very high). The distribution shows a strong relationship between election quality and the ability of observers to operate.
The general pattern is that, where elections have high integrity, domestic and international watchdogs are free to observe and report. In low-integrity contests, they typically face harsh restrictions.
The average score on both survey items is 2.2 (out of five). This suggests that restriction of election monitors is not among the worst problems of electoral integrity.
Unsurprisingly, PEI experts evaluated restrictions on both types of observers as unproblematic in high-integrity contests. These are concentrated in Western Europe and the OECD democracies. They are tightly grouped in the bottom-left corner of the graph, which suggests regional norm diffusion effects.
The variability among countries with medium electoral integrity is much larger. It ranges from almost no restrictions in Ghana or Moldova, with scores of about 1.5, to a quite adverse environment for observers in Singapore, with scores of about 3 on both variables. Singapore may be less susceptible to the sticks and carrots of shaming and international aid because of its independent sources of economic growth.
The span is even wider for low-integrity contests. Malawi and Guatemala allow observers almost free access despite running generally poor elections. Highly repressive countries such as Ethiopia, Belarus and Equatorial Guinea severely penalise observers.
Basically, election-related grievances are strong in all of these countries. But while Malawi and Guatemala guarantee rights of association, the closed autocracies at the upper-right corner of the graph severely repress any form of civil society engagement. Domestic observers are routinely jailed, harassed and denied access to polling places.
Thus, grievances and political opportunities both seem to be important drivers of observer access.
What’s happening in the outliers?
Iran is an interesting outlier. It is in the mid-range of electoral integrity. The PEI experts saw prohibitive restrictions on international monitors, but much freer domestic watchdogs.
Due to the country’s pariah status, no international aid spending provides supply-side factors for NGOs to monitor elections in Iran. More importantly, the legal framework does not provide for accreditation of international observers.
While independent citizen groups are not allowed, Iran’s Guardian Council and the General Inspection Organisation may field observers. This provides at least some domestic oversight. Given the regime’s demonstrated ability to quell even large-scale contention about fraudulent elections, it might be more willing to grant some access to domestic rather than international observers.
Conversely, Kazakhstan – another outlier – severely represses domestic NGOs while allowing some access to international watchdogs. The country, being part of the former Soviet space and a field of operations of the OSCE, faces stronger regional pressures to allow such access. However, by cracking down on domestic NGOs, the regime might be trying to deprive foreign observers of their primary information source.
In addition, the OSCE report on the 2015 election in Kazakhstan questioned the independence of a prominent observer group due to a lack of transparency in its funding. This suggests a strategy of closely managing some domestic observers and restricting independent ones.
Overall, grievances and political opportunities go a long way to explaining where and why election monitors are restricted. Both domestic and international observers face the most adverse environment and few opportunities to operate in countries where electoral contests have low integrity. These are the very places where their work may be most needed.
Observers enjoy a more conducive environment where elections already have high integrity. Despite such open access, often fewer domestic monitors are active in these countries. This may be simply because there are fewer grievances associated with poor electoral integrity.
Once more is known about both – the presence or absence of observers, and whether or not they are restricted – we can begin to unpack their possible impact.