As the Haligonian put it in the comments to this post:
One critical factor in this is that people are forced to make CPP contributions. It is not like a private plan that can be opted out of or at least can factor into an individual's employment choices. Gay people in 1985 simply had to make CPP contributions and they had no way of ensuring that the pension could be passed on to their partner.
If this is wrong today, which I think it is, then why wasn't it also a problem back in 1985? The people who lost out because of that are still alive in many cases and should be compensated because they had no alternatives.
Since my preferred rule would be to give the legislature a lot of leeway in relation to the transitional issues involved in implementing a constitutional decision, I have to answer this point. (So does the Court, which took the view that Parliament deserved no deference, but they should be able to determine the transitional issues on the basis of unworkable doctrine.)
The Haligonian's argument is seductive. Gays and lesbians contributed to CPP, so why shouldn't they get their money back?
Unfortunately, it is inherent in social insurance (and even private defined benefit or insurance schemes) that there is no direct relation between contribution and what you get back. If you want this, you end up with a Bush-style defined contribution scheme and -- ultimately -- libertarianism. CPP developed survivorship pensions because of modal heterosexual divisions of paid and unpaid work -- to protect women. Overall, it is not clear that same sex couples lost out financially for being treated as two individuals by the tax and transfer system. The reason for addressing the differential legal treatment between same sex and heterosexual couples was not that same-sex couples were disadvantaged financially. They weren't. The reason was that there was an obvious social message implied by the differential treatment, one that goes back to Judeo-Christian and perhaps naturally evolved attitudes that caused and cause grief for gays and lesbians.
So a retroactive remedy might well be unfair unless a lot of interconnected things were undone simulatenously. And doing that in regard to past transactions could cause a lot of unforseen hardship.
Presumably, these considerations were why the courts tended to suspend their declarations of invalidity when the original victories were won for same-sex couples.
There are other implications of prospective remedies:
--Precisely because they are less disruptive to society, they might encourage a more activist approach by the courts. On a traditional view of remedies, the constitution always said what it says now, and so governments ought to be able to anticipate an unfavourable ruling. Legislation is prospective because it is supposed to be about change and "progress". If judicial remedies are prospective, judicial thinking may become more legislative. I think that's what Andy was trying to get at here.
So from a judicial point-of-view, prospective remedies might mean more change through constitutional litigation.
--On the other hand, retroactive remedies create more incentives for clients and lawyers to engage in constitutional litigation. Kingstreet is going to mean a lot more constitutional tax cases. If Hislop had been an uncomplicated victory for the plaintiffs, then their lawyers would have become rich. And rich lawyers get imitated.