The research shows that scale is very important. Bigger scale means lower unit costs. Delivering pensions at low cost is tremendously important, because if a lot of the returns are eaten up in costs, then there are no pensions. So in that sense, scale is important.
As to whether it needs to be one national plan, the way I wrote it up, as the Canada supplementary pension plan, or whether you have a CPP-QSPP as a variant of that, or whether we have a number of regional plans, I think they can all achieve scale.
So now you get to the question of portability. Portability across the country, if you want labour mobility, is obviously very important. So even if you have somewhat different sponsorship, you want the formulas to be the same. You want them all to be target benefit plans. You want them all to have individual pension accounts, and you want all of the older workers and retirees to be part of an annuity balance sheet that is secure and will in fact pay pensions.
So it's an interesting question whether we end up with one national scheme or whether we end up with three or four regional schemes. I think that needs to be seen. I think we're going for regional schemes right now. I think what starts to happen, when people start to talk to each other about what they should do separately and where they should collaborate, is that you're going to find there are a lot of things that make sense to do together, and you may in fact end up with a national scheme that's not top-down but is in fact bottom-up.
The other question relates to compulsory versus voluntary. It turns out that the research showed the answer is neither, and that in fact an approach that automatically enrols workers in a scheme that then needs to be communicated is in fact cost-effective and works in their interest. The research shows that most workers will stay in a scheme like that. They will not opt out. However, if you have some libertarianism in you, then the idea of something that is compulsory, that you cannot opt out of, is somehow distasteful.
I think we can get to 95% of where we need to go by having an auto-enrolment scheme with an option to opt out.