The claim is actually the opposite: You can never make the claim that no regulation is good, because you don't know what you don't know.
(But you absolutely can still make that claim, in fairness; you just can't base it on a cost-benefit analysis from the customer's perspective.)
The OP said that this shows why regulation gets good results, and I challenge the evaluation by saying we have seen the benefit of this particular regulation, but not the cost.
You argue that is hard to know both the cost and the benefit of regulations. I argue that if you cant gauge it, then you can never conclude if its good or bad.
I rather put the responsibility of proving itself to the guys with the battons.
Here's where the trick is. I'm not saying "you can't gauge it"; you even acknowledged in the previous sentence that that's not what I'm saying. I'm saying it's complex, because in order to gauge it, you must have some of it, otherwise there is no way to produce reliable metrics.
Yes, that requires incurring some cost. But the risk of much higher costs comes down dramatically once the initial cost is paid, because the information it provides makes us able to avoid greater costs down the road.
What you're saying is that if we can't build a tower all in one go, we can't build it at all. I'm saying we can do it in reasonable stages, but we have to lay a strong foundation first.
It's not a fair comparison. The purpose of regulation in this case is to protect consumers of bank services - including other banks. Those costs are only calculable if you know the full scope of the bad behavior, which, without appropriate regulation ensuring reporting and review, you can't.
Of course, if you don't believe banks ought to be regulated in order to protect their customers, it doesn't matter. But that's a different conversation.