by Andrew Ellis
A vital task of democracy-building is understanding the links between national elections and the wider dynamics of democratic change, says Andrew Ellis.
Elections are a core part of the common understanding and practice of democracy. Yet experience – not least in countries emerging from violent conflict – increasingly suggests that electoral events conceived and held in isolation from their broader political context can become as much part of the prevailing political "problem" as their democratic "solution". This article reviews critical issues and challenges in the relationship between elections, popular validation and democratic consolidation.
In transition: the peace building element
Elections are considered as a defining and unavoidable element of any peace building process. They are clear, identifiable and newsworthy events. They are thus highly likely to be overemphasised by the international community, which has often regarded them as the benchmark point for its exit strategy from a peace building effort – getting out before international politics moves on to the next great cause and before domestic political and financial pressures grow to declare victory and go home. Local participants in negotiations – whether democrats or not – are fully aware of this.
Warlords believe that if they only wait long enough and make a few concessions, sooner or later there will be an election and the international presence will almost certainly be gone – as Charles Taylor, for example, made fully evident during and after his successful electoral campaign in Liberia in 1997. While electoral processes – elections and referendums – are integral to democracy, their context and conduct during transition requires much more careful thought than has sometimes been the case in the past.
Politics, negotiation, trust
Even when there is no rush to hold an election or a referendum, there never seems to be enough time for the organisation and implementation of the electoral process. This is no accident. It follows inevitably from the dynamics of negotiation, in which the value of political concessions is likely to be higher the later they are made, and the technical and logistic side of electoral planning is unlikely to be an important factor in the minds of those reaching political agreements.
Neither is any peace building or democracy-building process as a whole likely to succeed without a commitment to dialogue, local ownership and broad stakeholder inclusion and participation. Even when an inclusive process of dialogue and negotiation takes place, there is almost inevitably a deep lack of trust in electoral processes held in the context of peace building – which means that the inevitable errors and rough edges that are accepted as just that in established democracies often lead to suspicion and to damage to the credibility of electoral processes in transitions.
This can be just as true outside the direct peace building context. In the words of a former chief electoral officer of Guyana: "We have to deliver our elections to a higher standard than in Europe: people do not forgive small mistakes as inevitable, but automatically regard them as evidence of cheating or skulduggery".
Nonetheless, the challenge is to hold good enough elections rather than to set a one-off standard which cannot be sustainably repeated. There can be a "good enough" election in a politically flawed peace building process (as Iraq, for example in the local elections of January 2009, has shown): it is difficult to have a politically sound peacebuilding process without a "good enough" election sooner or later. In particular, the ability of armed groups excluded from a peace negotiation to disrupt the process tends to incline agreements towards the design of inclusive institutions which will involve all or most of these groups – although sometimes questions of transitional justice may temper who is allowed to participate.
The frameworks which then emerge tend to require the continuing construction and maintenance of coalitions after elections have taken place. However, the new politicians may find they need to acquire different and new skills to do this, especially if they have previously embraced the very different disciplines required in armed or insurgent groups – as has been borne out by the experience of Nepal.
Both peacebuilding and democracy-building are intrinsically political: any outcome inevitably advantages some stakeholders and disadvantages others, while actions that promote peace are not necessarily actions that assist democratisation – and vice-versa. A "good enough" election is about the political environment and conditions, not just about the technical competence and independence of the electoral administration.
Interventions which depend on importing external technical "experts" with ready-made "solutions" (often copies of the system used in the country of origin of the "expert") are particularly unlikely to be appropriate and may well be harmful. Electoral issues that are often thought of as technical but which always have political sensitivity include electoral-system design, franchise qualifications and evidence-requirements for registration, boundary-delimitation, absentee and external voting, and counting, tabulation and declaration of results.
Disputes, conflict, violence
Moreover, "good enough" electoral processes do not end when results are declared. The little things that go wrong often violate the individual rights of voters. When bigger things go wrong – or when little things go wrong and the result is close – there are wider political impacts: the results change, or are uncertain. Serious thinking about mechanisms for effective electoral justice is now taking place, with the independent judicial electoral courts developed in Latin America providing particularly interesting models.
Where electoral competition is itself the cause of violence, early-warning systems (looking at the context of the electoral process in each community and pinpointing likely trouble-spots) are important new tools; India and Colombia are among the leaders here. But this approach may be insufficient where the electoral process is not in itself the cause of violence but rather the catalyst for existing conflicts within a divided community – when different sources of advance intelligence become even more important.
It is especially easy for the electoral process to be a flashpoint when the community sees the police or the military themselves as a source of insecurity. Even where the popular view of the security forces is more positive, their world rarely intersects with that of the electoral community, and there is still a lot to do to familiarise police and military personnel with the behaviour that a democratic electoral process requires.
The growth of electoral independence
In designing and establishing electoral administration, a structure that enables its fearless independence is essential. Such independence means that the electoral administration does not bend to government, political or other partisan interests; though it is worth emphasising that the threat can come not just from overt political restriction or pressure, but from financial mechanisms which prevent the administration from accessing money and other resources when needed.
As democratic institutions have been established across the global south, many countries have followed India in establishing an independent electoral authority. While not all have matched the unchallenged respect with which the Election Commission of India is held or the success of other independent electoral commissions (such as that of South Africa), the model is now used by over half the world’s countries and territories.
Even in much of the francophone world, where a substantial body of jurisprudential opinion finds the concept of an independent electoral body untenable, a move towards independent electoral administration is visible.
Towards sustainable capacity
The key objective – building fearlessly independent electoral administration that becomes over time administratively and financially sustainable throughout the entire electoral cycle – has consequences for electoral assistance. This should be viewed as a long-term process. It should not primarily aim to support highly visible events and international observation exercises immediately surrounding polling day, but engage to support capacity-building over a series of electoral cycles.
However, electoral administration is only one of the claims on the limited financial and human resources that may be available within a country in transition. Organisational and staff development, and the building of institutional memory within electoral institutions, are essential. Generally accepted "best practice" has been developed since the mid-1980s in various knowledge resources (for example, handbooks and networks such as the ACE Electoral Knowledge Network and the world-standard electoral capacity-building curriculum, BRIDGE.
The use of new equipment and technology can sometimes be valuable, in particular for managing vast amounts of data such as the electoral register or the processing of the results, but the temptation to throw money at technology needs to be resisted (especially when this is done late in the day).
On the political side, the issue is whether voters, political parties and candidates will perceive the technology as a contribution to electoral integrity or as an unverifiable "black box" which increases the suspicion of fraud.
On the technical side, the needs for maintenance and for trained operating personnel, and the issue of potential obsolescence, should be considered. Commercial vendors of equipment and technology (for example of identity and registration systems) have their own interests; these are not the same as those required to build sustainable electoral administration.
The test of robustness
Sustainability takes time to build. Respect for electoral authorities develops only over a period of years, and their political robustness may only fully be proved in difficult circumstances. If election results are uncontroversial, then electoral-management weaknesses and failures will be less critical.
Indonesia’s independent electoral authority was less well organised in 2009 for the third legislative elections of the reform era than for the previous two, and substantial problems were visible with electoral registration in particular: but the overall results of the elections were clear and accepted. If they had been as close as the result of Mexico’s presidential elections in July 2006, with less than one-half of a percentage point separating the two leading candidates, the credibility of the elections could have been in the balance.
But in Mexico, while there were cries of fraud in the counting process, few such were made against the electoral register. The electoral management body was heavily challenged and emerged bruised, but its underlying strength was still intact.
Kenya’s elections in December 2007 offer a very clear example that political robustness is not the same as technical competence. The electoral authority had the technical knowledge and capacity for everything to work well – but it did not have the political robustness to cope with the political pressures of a disputed election (which may or may not even have been close). There have been few sadder recent electoral moments than the chair of Kenya’s electoral authority finally admitting that he had no idea who had really won.
Design, timing, sequencing
A challenge for peace-builders and democracy-builders alike is to facilitate choices in the design, timing and sequencing, implementation and assessment of electoral processes to a constituent assembly, legislative bodies and/or local government bodies – all of which are fundamentally political questions. The first universal elections in South Africa in 1994 took place after the process of transition had been under way for several years.
By contrast, the early elections held in Bosnia & Herzegovina in 1996 under the Dayton agreement led to the freezing into place of zero-sum identity politics in which the participants in the conflict played the major roles. It may thus not always be advisable to hold early elections for bodies that are not transitional in nature. But if holding early elections is undesirable, who holds power until elections do take place? When does democracy delayed become democracy denied?
Is it easier to encourage the beginnings of reconciliation by focusing initially on creating local-level institutions to engage in practical service delivery on the ground? This approach poses the question of who can legitimately hold national power, and be responsible for the design and resourcing of the local institutions, in the interim: in Kosovo, the United Nations mission did so, and elections were held at local level first.
Not all such approaches are as welcome: in Timor Leste (East Timor), the World Bank took an initiative to set up village committees as soon as possible after the conflict – but this action was clearly not welcomed by the UN transitional authority.
Representation, accountability, and the system
Election laws may be changed with alarming frequency as the weaknesses of a given electoral framework are discovered – or as parties seek to gain advantage of it. But the opposite can also be true: once laws and structures have been put in place, it can be difficult to remedy mistakes because the incumbent politicians and officials have vested interests. The provisions for amending electoral rules are another area which looks technical but is highly political in its effect.
Electoral system design is a key political choice, both in assembling the electoral framework and more generally in the institution-building process. Discussions about electoral systems and the design of legislatures talk a great deal about representation and accountability, but both words have multiple meanings. The basis chosen for representation and accountability is an important factor in the incentives for accommodative or "winner take all" behaviour by those who hold or gain power.
Do elected legislators respond primarily to the whole electorate, all voters, party supporters, party members, party activists, party leaders, or whoever is going to give them their next job? Term-limits are one factor: in addition, some constellations of electoral systems and institutional frameworks are intrinsically more likely than others to promote turnover of elected members.
This relationship is not simple: well over 90% of incumbents are re-elected to the United States’s House of Representatives using a first-past-the-post electoral system; but the same system as used in Papua New Guinea (PNG) until recently produced a turnover rate closer to 50%. The incentives for the PNG members to take benefits from their position while they were in a position to do so were self-evident – as were the consequences for the coherence of the PNG parliament.
Electoral frameworks do not only create incentives for candidates: they create incentives for political parties, who measure using different criteria. When a voter is invited to express preferences between candidates, she or he may well prioritise according to how close their positions are to those of voters. A political party, however, gains strength by attracting new voters, who are likely to be supporters of parties which are relatively close: it has an interest in the failure of adjacent parties in the political spectrum.
This is demonstrated in the recent history of Fiji: the unsatisfactory nature of the 1997 institutional framework in approaching this question has played a substantial role in the events that have led to the full military takeover.
Electoral-framework debates are also conditioned by a conventional ideal which aims to establish a stable political party system based on programmatic differences between parties. However, this framework is often an unattainable ideal in the age of mass political communication.
Today’s new challenge – even more difficult than reaching the ideal – is how to entrench service-delivery and accountability in a volatile political party system, in which voter choice may well be based on leadership and/or identity rather than on programmatic difference. The real danger should perhaps not be seen as leadership or identity themselves, but as the entrenchment of zero-sum politics.
Electoral processes: the big picture
Many lessons have been learnt since the 1970s-1980s about approaches to electoral processes which are more likely to constitute comparative good practice; this article notes only some of the most important. But the more experience, expertise and analysis – from the global south as well as the north – that exists on electoral processes in democratisation, the more complex and context-driven they turn out to be.
Democracies are more likely to be stable than authoritarian states: democratising societies may not be. Peacekeeping activities may provide space for peace, but do not necessarily assist democratisation processes in the longer term. Electoral processes are an essential element of democratic change, consolidation and stability: but in the early stages of transition in particular, they can be flashpoints with the potential to encourage the re-emergence of conflict, and if badly designed, can entrench forces that do not promote democratisation.
Every element of a process of change – in the sphere of democracy-building, electoral processes, constitutions, political parties, legislatures, decentralisation/devolution, independent judiciaries, among others – is likely to be linked to every other. The democratic nature of the institutional framework as a whole cannot be separated from its ability to deliver development and services; nor from the space allowed for democracy as a whole (by, for example, security-sector issues).
Both democracy-builders and peacebuilders will always need to engage with electoral processes in depth – the devil is certainly in their detail. But it is critical that they never lose sight of the inextricable links, potentially both positive and negative, between electoral processes and the wider process of democratic change.