Evaluation Report of the New Methods of Voting that were Used during the Municipal Elections of November 2005

Good morning to you all.

Thank you for taking time to be here today… two weeks from the first anniversary of the municipal elections that, as we remember, were rather “eventful.”

“Eventful” because the voting itself and the dissemination of results were marked by what can be referred to as “disturbing failures” related to the use of new methods of voting.

The municipal elections of November 6, 2005, were the first to be held at the same time in all Québec municipalities. It was also the first time that so many municipalities were using electronic voting or the postal ballot to enable electors exercise their voting right.

We all remember what happened that day: not only did the electronic voting systems fail, but the corrective measures applied were insufficient, poorly adapted and often late.

Result: in many places of voting, the returning officers had to deal with major technical problems… First consequence: voting was delayed … Another consequence: in many cases, the release of results was also delayed whereas speed was one of the intended benefits.

But something else of a more serious nature happened that day: the confidence of many actors and more specifically the confidence of electors in the democratic process of municipal elections, was eroded. Not only were we wondering what may have happened …. but it was unbelievable that an exercise that usually takes pace in serenity could turn, in certain places, into such confusion. I must confess that on November 7, 2005, I was not more informed than you. As far as the organization of municipal elections is concerned, the law states that the primary person in charge is a municipal official, that is, the returning officer, with my organization being mainly a “support partner.”

With this in mind, on November 7, 2005, I can confirm that at the Chief Electoral Officer, we were also in shock and I was able to publicly express my concerns that day.

Some people would have liked a permanent halt to all testing of new methods of voting. For my part, I wanted to know what happened before drawing conclusions, which is why I created an evaluation committee that conducted a review of new methods of voting that is unprecedented in Québec. Formally, responsibility for conducting such a review did not fall on the Chief Electoral Officer. We, however, decided to conduct the review because we were best placed to do so.

This is why close to a year later, I am once more before you following the completion of our evaluation work.

I would like to start by sharing with you our findings regarding what happened last November given my commitment to shed full light on this issue … The report that you have before you contains most of the answers that everyone was awaiting on the causes of the problems that we experienced.

I am also coming before you to say that electronic voting systems, as they were used, are more troubling than we would have thought after the last municipal elections. And it is possible that we would have experienced even worse problems than that was the case on November 6, 2005.

As you know, we have been using electronic voting for ten years with no major problem… Ten years of increasing satisfaction by municipalities that kept asking for it… All of which gave some credibility to this new manner of holding elections. What we experienced on November 6, 2005, and what our review of the situation revealed is that this approach was more hazardous than we could have imagined.

My decision following last year’s municipal elections was to apply a moratorium on all new requests to use new methods of voting. What I recommend today is that this moratorium be extended until the legislator decides whether or not to maintain or use of the new methods of voting in municipal elections. And in case it is maintained, corrective measures must be implemented.

But before providing you further details, permit me to briefly describe the work we did.

Following last year’s municipal elections, I appointed staff who, over the past year, have concentrated on “the case” of the elections of November 6 and on the problems that ensued.

Those primarily responsible for these elections were the returning officers who organized them. We met most of them after analyzing the 144 evaluation reports that they sent to us as required by law.

We also met the three suppliers of electronic voting systems and the supplier of the postal ballot, in addition to reviewing the written reports that they produced. After our investigations we also reviewed the complaints received. You probably remember those of Pierre Bourque in Montréal and Hugo Lépine in Québec City. We reviewed the motions for recounts or new compilations filed in court as well as the decisions rendered by the courts.

You would remember that there was a high number of rejected ballots in certain municipalities. The evaluation committee visited each of the seven municipalities where the total rejection rate was 4% or above. It reviewed, a posteriori, each of the situations that were previously examined by the returning officers, at the request of the Chief Electoral Officer.

Furthermore, we conducted a review that had never been done before in Québec and probably in very few places in the world: we conducted technical audits and testing of the seven electronic voting systems used by the municipalities last November 6th. For this purpose, we required the expertise of the Centre de recherche informatique de Montréal (CRIM).

Another important remark: the primary objective of our evaluation was not to identify who was responsible for these problems since all the actors in the 2005 municipal elections must share some responsibility. We did not evaluate people but systems and processes… in addition to see how the various actors were involved. Understanding that in all these matters, I am convinced that all the actors acted in good faith.

What were the results of all these meetings, all these reviews, all these audits and all these tests?

I will start with what, we believe, caused the problems that occurred during the municipal elections. The following points are worth noting regarding the problems encountered during the municipal elections:

  • An imprecise legislative and administrative framework that did not adequately assign roles and responsibilities or address the risks inherent in electronic voting.
  • Absence of technical specifications, norms and standards that would have guaranteed the quality and the security of the voting systems used;
  • Poor management of the voting systems, especially inadequate security measures that encouraged errors and contingencies.
  • Absence of solutions or insufficient solutions in case of problems… the famous “plans B” that would have been so useful.

More specifically, it is possible to pinpoint a number of circumstances that increased the risks:

  • Inadequate testing of voting machines, inadequate quality control of components and inadequate measures to ensure the security of the methods of voting and thus, the integrity of the vote.
  • Certain returning officers delegated too much responsibility to certain service providers.
  • Support and troubleshooting services were sometimes inefficient and inoperative.
  • There was lack of transparency in information provided to candidates and their representatives.
  • Processes regarding the use of voting systems were not documented.
  • In addition to managing election processes, returning officers had to manage the technical aspects and the services offered by “suppliers of technical services.” In reality, they had difficulty harmonizing their responsibilities with those of suppliers leading, for example, to loopholes regarding the training of election staff.
  • As I mentioned earlier, our primary goal is not to blame one supplier rather than another … But we should note, nevertheless, that one of the suppliers overestimated its ability to simultaneously serve a large number of municipalities, particularly the largest ones.
  • I would add that, in our opinion, if other suppliers had signed as many contracts with as many municipalities, they would probably have experienced the same problems because they ran the same risks due to the very nature of all the voting systems used.
  • Another circumstance that increased the risk factors: imprecise contracts and incomplete specifications blurred the relationships between the municipalities and their service providers.
  • There was also the absence of an independent expert who was specialized in electronic voting to which returning officers could have turned.

Those then are a synopsis of the reasons for the difficulties that we experienced during the municipal elections of November 6, 2005.

But as I mentioned, we went further than simply diagnosing the elections of last November. We wanted to conduct an in-depth analysis of electronic voting systems.

For this, we did an audit and conducted technical tests on the seven electronic voting systems used on November 6, 2005. These were made up of three systems of electronic ballot boxes, two voting terminal systems and two types of computerized lists of electors.

The result of these audits leads me to recommend very stringent conditions for the future use of electronic voting if, of course, the legislator decides that it can be used again in municipal elections in Québec.

I would like to remind you that the Act Respecting Elections and Referendums in Municipalities still enables municipalities to use electronic voting after signing a memorandum of understanding with the Minister of Municipal Affairs and Regions and the Chief Electoral Officer. The moratorium that I am applying now does not modify the law, but my agreement is required.

What I am submitting to the Legislator for reflection is firstly that electronic voting does not seem to have lived up to its promises in many cases in November 2005. Among others, in certain municipalities, the results were not known quickly … rather, the reverse occurred. Furthermore, according to the returning officers who were consulted, electronic voting did not lead to the expected savings: it cost as much, if not more, than the traditional method of voting.

I am also reporting to the Legislator that electronic ballots and voting terminals as they are programmed and used, are vulnerable technologies.

These systems present major risks related to the fact that formal protection and security measures are insufficient or absent. They are exposed to major service or network breakdowns and are vulnerable to technological attacks.

Our review of the manner in which electronic ballot boxes are tested, installed and managed during an election also resulted in very troubling findings: there is a lack of knowledge of the components of voting systems. Furthermore, the actors during elections – including the services providers – lacked expertise regarding voting systems.

At a time when we are having confidence in technology in many situations in our daily lives, we can imagine that one day, electronic systems could assist us in holding an election.

But for this to be done in Québec, given what we experienced last November 6th, we would have to perform an in-depth look at the manner in which we use electronic voting systems:

  • The legislation governing the use of these methods of voting must be reviewed and better defined, including the memoranda of understanding signed by the municipalities with the Chief Electoral Officer and the Minister of Municipal Affairs and Regions.
  • Rigorous technical specifications as well as security and reliability norms and standards must be adopted before any future use of a new method of voting. A group of experts must be created in this respect.
  • An independent authority must be vested with a mandate to monitor and the powers to audit and control the norms and standards related to the new methods of voting. If you say that the Chief Electoral Officer could be this “independent authority,” I would answer that we also need to “develop an expertise” and that we should be given the considerable resources required to play this role. I am not sure, however, if an election administrator should play the role of “approver of electronic devices.”

More specifically, we recommend that if electronic voting is maintained, a supplier’s ability to offer services to several municipalities should be verified.

The municipalities should also sign more stringent contracts containing more precise specifications regarding the professional services required and the voting systems used.

Regarding the actors in a municipal election, the role and responsibilities of the Chief Electoral Officer, the Ministry of the Municipal Affairs, returning officers and service providers need to be clarified.

Finally, as far as security is concerned, measures should be adopted so as to guarantee the integrity of the electoral process. For instance:

  • Competent authorities should have access to the programming codes and source codes of the software and voting systems used.
  • The implementation of mandatory and complete tests on all equipment to be used in an election. In computer security, the word “test” is probably the key word … or the most important “password”.
  • The establishment of backup plans covering all potential problems.
  • Measures aimed at ensuring the transparency of election operations should be envisaged.
  • All those responsible for programming and installing systems and software and providing technical support and troubleshooting should take an oath.
  • Strict measures should be implemented for the storage and safekeeping of systems used.
  • Measures should be adopted to ensure that, after the election, the supplier destroys data recorded on the electronic voting systems.

A word on the postal ballot… The postal ballot went relatively well in all the 22 municipalities that used it last November. We included it anyway in our evaluation of the new methods of voting… because it is one of them.

It is worth noting that the postal ballot can be very useful for electors, particularly those who are not domiciled on the territory of a municipality. I am thinking here of owners of chalets.

In this regard, we recommend that returning officers adopt a model postal ballot similar to the one used for voting by electors outside Québec and voting by inmates.

Thus, the systematic sending of voting kits to all electors domiciled in a municipality should be stopped: electors should request them. And systematic solicitation, as was the case in one municipality, should be prohibited. In conclusion…

I believe that following the findings that are presented in our report, we have to question ourselves on the added value of electronic voting. And we should also be very aware of all that should be put in place in the form of norms, controls, audit procedures, training, etc. for electronic voting systems to be trusted again.

Thank you for your attention; I am now ready to answer your questions…