Conventional wisdom holds that ever-evolving technology can improve efficiency and performance in most areas of human activity. As the UN’s electoral focal point, the Department of Political Affairs is familiar with the introduction of technology in electoral processes, for example. Elections and technology are, on the surface, a perfect fit. After all, if there is one thing computers, advanced software and the Internet do well is render the use of paper documents like ballots and voter lists obsolete. There has been widespread belief that information and communication technology (ICT)would bring about revolutionary improvements in how polls were conducted, and all at a lower cost. In light of recent experience in different parts of the world, the faith in technology as an electoral panacea has faded. At a recent meeting in DPA, electoral experts explored why the marriage of high tech and elections has not been completely smooth, and what can be done about it.
Countries worldwide are increasingly relying on information and communication technologies in the conduct of electoral processes, moved by a desire to improve the accuracy, security and integrity of elections. This trend has been particularly noticeable regarding voter registration, results management and, more recently, electronic voting. In some cases, the introduction of technologies is envisioned as part of a long-term plan for improving efficiency or lowering costs. In other cases, automatization is prompted by a sense of urgency to overcome a specific obstacle.
If countries so request, the UN, as well as other organizations such as IFES, can help Member States make informed decisions that could bring about consensus and public confidence over the use of technology in elections. Experience has shown that new technologies, if not properly introduced, can present challenges in terms of the trust in the electoral process, potentially creating bigger problems than those they were intended to resolve. In a recent report to the General Assembly, the Secretary-General highlighted the importance of taking politically and financially sustainable decisions about technology, and noted that technology does not, in and of itself, create confidence or prevent fraud.
How, then, can new technologies be best applied in elections? That’s the question three international experts came together to discuss on 4 October in the Department of Political Affairs (DPA). Peter Erben, Senior Global Electoral Adviser at the International Foundation for Electoral Systems (IFES), Niall McCann, Policy Advisor on Electoral Assistance at the United Nations Development Programme (UNDP) and Simon-Pierre Nanitelamio, Deputy Director of the Electoral Assistance Division of the Department of Political Affairs, drew on lessons from more than two decades of support to electoral processes by the United Nations and IFES.
“The relationship between the success of an election and the use of technology is not always straightforward,” said Nanitelamio. “New technologies cannot, by themselves, build trust in an electoral process, and should not be seen as a technical panacea to electoral problems that are fundamentally political in nature. One cannot rely on ICTs to solve structural issues affecting elections.”
Nanitelamio said that the UN does not consider the use of technology in the field of elections an end in itself. Rather, technology should be used as a tool at the service of electoral processes to address a specific problem: Is there an issue with the accuracy or credibility of the voter register? Is there an issue with the transmission or the management of the electoral results? Is there a problem with the voting or the votes tabulation procedures?
UNDP’s McCann pointed to a specific example of the use of technology in elections, specifically in relation to voter registration. Some countries that do not draw their voter register directly from civil registers but instead have a stand-alone and active registration process for elections have embraced biometric technology to compile digital lists that are “de-duplicated” to ensure that each voter has registered only once, helping preserve the “one person, one vote” principle. However, biometric voter registration systems are coming under increasing criticism from wider public administration and population registration experts over the lost opportunity of focusing such human, financial and technology resources on one functional register of the population. Such resources, it has been argued in recent research by the World Bank and the Center for Global Development, among others, would be better used solving the very issue that leads to the need to detect multiple registrants in the first place – the under-development of the foundational civil registration and wider identity management system in the country.
“Ensuring a holistic, digital birth-to-death population registration system with inter-operability between various state registers while founded on a solid data protection legislative framework would allow the benefits of biometric technology to empower citizens to access many additional public services, not just voting,” said McCann. Enormous savings benefits could be derived by state agencies not having to carry out parallel mass registration exercises that invariably underachieve full coverage as they target only specific audiences.
Cost effectiveness, a major argument for the introduction of electoral technology, is not always a given, Nanitelamio pointed out. “Technology requires significant initial costs and subsequent costs for storage and maintenance of information technology(IT)equipment, which often has a limited lifespan and needs to regularly be updated or replaced,” he said. “ Sustainability is almost impossible to achieve where inappropriate technologies are implemented. The IT solution should be re-usable and able to be sustained locally without relying on external experts, technicians and vendors.”
Erben recalled that technologies for voter registration, results management and electronic voting have proven to be vulnerable to failure and security breaches, distrust by contestants and voters, inflated costs and legal challenges. Nanitelamio added that recent elections have highlighted the fact that, while it is difficult to collectively hack voting machines or results management and transmission systems, the IT solutions remain vulnerable to hacking. “In such cases, solutions connecting voting machines to the internet needlessly creates another security weakness.” he said. “This is especially dangerous when those machines don’t create a paper trail that allows for double-checking or auditing.”
IFES argues that the combination of paper-based electoral systems and electronic systems can lead to good results and can address mistrust. “Traditional paper-based systems and electronic systems, used together, mutually reinforce each other, leveraging the significant symbiosis between the old and the new,” Erben said.
“There’s no ‘one size fits all’ solution,” added Nanitelamio. He argued that if dealt with adequately, the benefits of introducing technology in the electoral process could far outweigh its numerous challenges. When deciding which technology to use in elections, the challenges need to be carefully considered and balanced against anticipated benefits. The relevance of each possible benefit and disadvantage will vary from country to country, as will the challenges and issues. Any proposed ICT solution should therefore be tailored to the needs and the specific context of each country. As there are many considerations to be taken into account, the appropriateness of a solution will vary from one context to another. It is therefore advisable to conduct an inclusive feasibility study involving consultation with all relevant stakeholders. The feasibility study should assess whether the introduction of technology is feasible throughout the country and within the proposed timeframe and should cover areas such as legal requirements, country’s infrastructure, internet availability, computer literacy, training needs, funding, cost-effectiveness and sustainability.