The discourse around electronic voting machines (EVMs) is heavily politicized. Stakeholders have taken up diametrically opposite positions. Proponents believe EVMs will fix structural issues in Pakistan’s elections and restore trust in a fundamentally broken process. On other hand, opponents contend that EVMs do not inspire confidence.
They can easily be hacked to manipulate election results. They have been phased out in technologically advanced countries, and the deployment costs are expected to be astronomical. These points are valid, but they only tell half the story, and it is little wonder the mainstream discourse has stalled.
I was part of a team at NUST that recently concluded a research project aimed at depolarizing the EVM debate. This project, funded by the Pakistan Institute of Development Economics RASTA grants program, addressed popular misconceptions around EVMs and provided a rigorous framework to ground the discourse and spell out a roadmap to deploy EVMs in Pakistan.
We believe that the EVM debate, contentious as it is, can be resolved in a way that highlights the positive automation benefits of EVMs and minimizes the security issues, costs, and other negatives. EVMs can be made to work for Pakistan, provided that we address various critical research gaps in our understanding of these machines.
And here we encounter an unfortunate and inescapable truth: whenever election technology – be it EVMs, Internet voting, or result transmission systems – are deployed without the necessary homework and due diligence, these systems are likely to fail. This results in costly and internationally embarrassing mistakes and risks undermining vital trust in election results and government for years to come. This point cannot be emphasized enough.
We witnessed this ourselves in 2018 with the failure of the Result Transmission System (RTS) at the critical hour on the eve of the election. The RTS was hastily deployed without any transparency features or adequate pilot runs. Likewise, the Internet voting system developed indigenously for overseas Pakistanis in 2018 featured structural and elementary problems and failed security audits twice. There had been no homework or study of international best practices in this domain.
Read more: Imran Khan’s exit: The day democracy died in Pakistan
The EVM landscape, too, has to be navigated with great care and effort. This is because, whereas EVMs may be very simple machines, they exist in a complex ecosystem with a wide variety of different factors at play, including political, cultural, legal, technological, financial, and logistical considerations.
EVMs have to be designed so that they are easily usable and do not inadvertently marginalize certain socio-economic groups. EVMs have to be transparent and auditable or supported by rigorous procedural checks and balances, which is extremely difficult in developing countries where systemic corruption is. A rigorous legal framework has to be developed for EVMs and effective dispute resolution mechanisms.
Moreover, EVMs have to be usable for at least 15-20 years to justify the investment. They will be stored, policed, transported, handled, repaired, and recycled at significant extra costs. Indeed, the costs of building the necessary ecosystem around EVMs in terms of logistics and manpower for a nationwide deployment might significantly exceed the cost of procuring the machines themselves.
Our mainstream discourse on EVMs neither recognizes these challenges nor acknowledges our current lack of expertise and capacity to undertake this task. This is an unfortunate trend observed and documented in various developing countries, a tendency to fetishize election technology to attribute extreme powers (or dangers) to it without actual research on the ground.
In our final project report, we offer several recommendations derived from international best practices to bridge the gaps in our understanding of EVMs. Here I summarize three strategies that are vital to overcoming the impasse in the national discourse on EVMs, forging a national consensus, and radically improving trust and transparency in our elections:
Use novel new technology to make EVMs transparent & auditable
The reason EVMs have a long history of negative press is that they are proverbial black boxes, i.e., votes are input into them, they output the results, but what happens in between these two steps, the inner workings of these machines are not transparent to us. International researchers have repeatedly demonstrated how easy it is to tweak the software on EVMs to manipulate votes in bulk. There are documented accounts and videos of EVMs flipping votes and inexplicable glitches affecting totals and swinging election results. There are even tutorials on how to manipulate EVMs on the Internet.
This is why various technologically advanced countries, including the Netherlands, Ireland, Germany, and Japan, have phased out EVMs. The sanctity of the vote takes precedence in these regions. In 2009, the Federal Constitutional Court of Germany effectively terminated Germany’s EVM experiment with the observation that an EVM “only meets the constitutional requirements if the essential steps of the voting and the ascertainment of the result can be examined reliably and without any specialist knowledge of the subject … The very wide-reaching effect of possible errors of the voting machines or of deliberate electoral fraud makes special precautions necessary to safeguard the principle of the public nature of elections.”
Read more: Why electronic voting machines cannot be used in elections?
Other countries opted to attach voter-verified paper trails (VVPATs) to EVMs as an independent check to verify they work correctly. But whereas paper trails do alleviate some concerns, they are increasingly being questioned nowadays. In India, several leading political parties have contended that the EVM paper trail is largely “ornamental,” and rigorous checks are needed to make it effective. Moreover, preserving the chain of custody of the paper trail and auditing it effectively is no easy feat in developing countries, where, we emphasize again, corruption is not isolated but systemic.
In recent years, the good news is that there have been revolutionary advancements in election technology that empower voters to audit the workings of these machines to their satisfaction. These include the development of risk-limiting audits, which are very efficient and effective counter-strategies to verify a small sample of the paper trail that gives citizens confidence in the results reported by the EVM.
Moreover, some of the world’s leading information security researchers have developed cryptographic solutions which allow voters to double-check all critical operations of an election. At the time of vote-casting, voters can opt to get a cryptographic receipt from the EVM with which they can ensure that votes were cast as they intended, received by the election body correctly, and the final result was correctly computed.
These solutions give us unprecedented transparency into the workings of EVMs and conclusively resolve the security apprehensions raised about EVMs. Several regions have started piloting and deploying these solutions, including Estonia, Australia, and several US states, whereas civil society groups have raised calls to introduce them in India. It is critically important that we also take advantage of these novel technologies to secure EVMs.
Undertake multiple rigorous pilot projects & phase deployment
There are many types of EVMs available on the market, and pilots are essential to determine which type and features are best suited to ground realities in Pakistan. These pilots have to be in a mix of urban and rural areas to ensure that an adequate cross-section of the electorate is represented in the results. Multiple pilots, which simulate the entire election cycle from EVM procurement to configuration, staff training, logistics, and result announcement, will also highlight the distinct pros and cons of different EVM models in a way that a simple study cannot.
Read more: Overseas Pakistanis: Hail, the new Kingmakers
This is why international best practices generally recommend that new election technology be deployed in a phased manner, starting with small-scale pilots, typically in non-political elections (like university bodies, trader organizations, councils, etc.), and then graduate to politically binding elections, which slowly expand in scope and size. This is a constant evolution, and with every iteration, there is an opportunity to learn and adapt and improve the machines and the processes. This is the route taken by success stories, including Brazil and India, which took over a decade to scale up to nationwide EVM usage.
Engage in extensive stakeholder consultation
Another prominent finding in the research literature is that those countries fare significantly better with election technology where the election authorities are relatively independent, and the election technology debate is democratized. Getting all stakeholders on board every step ensures that all voices are heard and that skeptical concerns are fully resolved.
Overall, this inspires greater confidence in the technology itself. If this debate plays out in public, the citizenry benefits from this exchange of views, leading to greater confidence in the process. Moreover, if any mishap occurs after EVMs are deployed, stakeholders will be invested in resolving the issue in a transparent manner rather than blaming each other.
To action these recommendations effectively, we suggest in our report that ECP urgently invest in an in-house Research and Development (R&D) cell to build study the application of new election technologies and design and oversee rigorous pilots of EVMs. We also suggest ECP, as custodian of elections in Pakistan, take the lead in engaging stakeholders – government, opposition, civil society, technologists, media, and activists – in a grand national dialogue to harmonize different views, resolve differences, and forge a consensus on the use of EVMs. In this matter, we would quote US activist and civil rights leader Bayard Rustin, “If we desire a society that is democratic, then democracy must become a means as well as an end.”
Read more: Electoral reforms – put the last first
The future for EVMs in Pakistan has never been brighter.
Dr. Taha Ali teaches in NUST, has a postdoctorate from Newcastle University in election systems, and advises the government and ECP on electronic voting. He can be reached at firstname.lastname@example.org