Okay, I'm back from holiday now.
Finally got around to seeing this question.
I don't know the actual tire dimensions on the new C5 bike, so I have to guess.
I just measured a stock 19" wheel with an Avon SM stock tire on it, and it comes out to about 26" diameter. That's a rolling circumference about 81.68".
Assuming that the new tire has a profile of the same 3.5" profile that the Avon SM has, then by reducing the diameter of the wheel by 1" would make it 25" diameter, with a rolling circumference of about 78.54".
Divide the smaller into the larger, and we get the smaller circumference being about 96% of the larger one.
So, by induction, we can estimate that the newer 18" diameter wheel/tire system will provide 96% of the older 19" one. This means the gearing will be about 4% shorter on the bikes with the smaller wheel.
I say "about" because some assumptions are being made here, which I'm not certain are correct, about the tire profile.
But, it's in the ballpark.
For reference, this would be about in the neighborhood of one more tooth on the rear sprocket, or about half a tooth less on the front sprocket.
Not a huge amount, but it's something.
However, we are not apprised of the engine primary sprocket ratio, which could affect the overall gearing ratios, and they may have different primary gearing in the new UCE that accounts for this rear wheel circumference. I don't know.
If there are numbers available which state "speeds in gears" at specific rpms of the engine, then we can have more data to go on.
If there is nothing else different, then the 18" wheel will spin up a little faster for acceleration purposes, and have a slightly lower top speed than the same bike with a 19" wheel. But, it won't be a very big difference. Probably about 3mph less at the top speed end. Better for hill climbing and acceleration, because of more mechanical advantage, but sacrificing a small bit in top speed.
As with our other bikes, a front sprocket change could give you gearing changes of a wide selection, so I don't really see a problem to it, as long as they made some adjustment to the speedo head to account for it at the gauge. As our speedos are already reading at about 10% too fast, I don't think any further variance away from true speed would be acceptable. They need to get a handle on this speedo thing anyway, and now would be a good time to make a speedo that actually tells you your speed.