If you say that there will be a measurable improvement in safety, I'll buy it.
Sorry Pilt, that's exactly what I wasn't trying to say. Perhaps it was the wine affecting the clarity of my post - and I'm another glass an a bit into the evening now so I hope this is intelligible!
I don't think you can measure safety in in this way. AIRPROXs and the like are a very coarse measure and, anyway, you would not be comparing like with like (i.e. CAS with FIS with different expectations of whether or not you'll meet another aircraft). But the very fact that aircraft will be flying in a known traffic-traffic environment, where all other aircraft will be under surveillance (or, at least, known about) and where traffic information based on stated intentions can be given means that the risk of collision is reduced - hence, it is safer.
How much safer - who knows? The big sky theory works pretty well already - the probability of collision is very small (marginal in statistical terms). The delta because you're in CAS can be calculated but it's a fairly meaningless figure because it is going to be based on all manner of assumptions. But if the assumptions are anything close to valid I can't believe that the delta would be negative.
landed, I agree. And the point I was trying to make was that you can't do much about the airspace - so perhaps one's effort would be more usefully expended on putting pressure on the CAA to put proper supervision mechanisms in place to ensure that equal access is assured. I seem to recall some time ago there was some sort of reporting system whereby pilots could submit a report if they believed they were unreasonably denied access to CAS. I don't know if it was a permanent scheme or a survey or whatever but maybe it would be worth investigating further to see if the possibility still exists.