There is certainly a range views on that issue. I can speak of our experience.
First of all, we need to not forget that we still do enumeration during an election campaign. We do it in a targeted fashion, in neighbourhoods with high mobility and all these things. We do targeted revision on reserves, for example. There is still a form of enumeration taking place for exactly the reasons you mentioned—people are highly mobile or hard to reach—so we do enumeration there.
One of the reasons—and think my colleagues across the country at the provincial level would agree—is that it's increasingly difficult to reach out to people through enumeration. We often knock at doors, but nobody answers. People are busy. They have different schedules. It's a challenge to recruit workers to do that kind of work. Even StatsCan is moving away from those surveys. The other thing—and it's unfortunate—is that in many cases, we can't find staff, for security reasons, who would go into certain neighbourhoods, yet they're probably the neighbourhoods that would most benefit.
As an alternative, we have the permanent register. What we did last time around also was launch our online registration service. That allows any Canadian at any time, at their convenience, to add themselves to the register or change their address, for example, if they've just moved recently.
Of course, looking at it from a cost perspective, the national register as a permanent list is much cheaper than enumeration. Yes, we lose contact with electors. Perhaps one way to offset that is to beef up the civic education program that exists.
I just want to take up one point. You inquired about the cost, which is absolutely legitimate. Of course, many of those recommendations don't bring any additional costs. Some of them bring extra costs, of course; an example is opening advance polls for longer periods of time. We estimate it's probably $500,000 an hour to have those polls open. With what we're proposing, we're talking about $6 million per election.
Civic education, registering youth at 16 or 17, would also have a cost. It's very preliminary, but we estimate that it's between $5 million and $10 million. We estimate that the first round would be more expensive. When the system stabilized, it would be a regular program, and the costs would come down. These would be the larger cost increases caused by those recommendations.
Of course, I can't speak for modernization. Now technology at the poll will be something else, but until we have devised a system, it would be premature to talk about it.