Prior to compliance legislation (like PCI, GLBA, HIPAA, SarBox, etc.) there wasn't much downside to having insecure systems; or more accurately, the downside only hit you after something bad had happened, so for most organizations it didn't make sense to worry about it until that point.
In today's business climate, the costs have increased somewhat: Potentially losing the ability to process card transactions, having to do breach notifications and pay for credit reports, being fined by the SEC, being made fun of on messageboards, etc.
They're still mostly only applicable after something bad happens though. It's like trying to convince people to back up their hard drives. There's a percentage of them who fundamentally accept that it's a good idea, and are proactive about it. There's probably a larger percentage who only consider it after they've had a drive failure and lost a bunch of data. I don't think there's anything you could say to them to make them understand that though.
But the "good" news (and I use that term loosely) is that the costs of "non-compliance" are going up, and I don't just mean from a regulatory perspective. Sony will spend a lot of money dealing with the implications of the PSN hacks. Citibank will as well.
Executives of other companies who weren't directly part of a breach will sit around in conference rooms and ask "So, what happened to them couldn't happen to us, right?" and some CSO will try to stifle a laugh and say "Oh no, we'd be screwed if the same thing happened to us."
People keep asking if good security is even practical. Security experts know it is, but nobody else does. If the incentives you describe were actually working, we would be properly assigning blame to bad security, instead of everyone simply asking if good security is even possible.
My point was that they exist, and believe it or not, as bad as they seem now, they are better than they were when I first started working in InfoSec ten years ago.
I'm not a "visionary" type person, but the only incentive that I can think of that would actually have practical implications for the consumer would be the ability to hold a provider responsible for bad security decisions (ie: Your site was hacked, my information was compromised, so now I'm suing you).
I also don't think this will ever happen (on a large enough scale to be effective); for a variety of reasons.
I know this sounds stupidly obvious, but companies on the whole will only take security seriously when the cost from being insecure is greater than the cost it will take them to secure things.
Maybe insurance will be the thing to finally tip the scale. Companies are already buying "hacker insurance", and I imagine if there's any industry with an incentive to enforce best practices, it would be the insurance industry.