OK - I thought it was in there - but I just scanned and didn't see it either! Guess I need to do some edits.
Anyways - I believe I originally got this from Elk's web-site, but can't find it now - so here's the basic gist:
UPB is a peer-to-peer technology. There is no master controller; each device is capable of sending links, and each device is capable of responding to up to 16 different links. The problem with the status is that I can have a link that's "All Off" which in theory could affect up to the max 250 devices. When you activate a link from a switch, that switch has no idea how many others are programmed to respond to that link; likewise, each switch that responds knows what it's supposed to do, but has no idea how many others are supposed to respond as well. If they were to all try to send their status after an activation, with say even 20+ devices, there'd just be no way for them to all communicate at once; they'd just pollute the power lines and no reliable signals would get through. UPB as a protocol doesn't have much in the way of collision prevention. It's for this reason that no switch will send its status after it has responded to a link.
Many of the better controllers have gotten around this by learning which switches respond to a link, then polling them individually after they see a link activate; I know Elve does this; and HAI does this via their own HLC protocol. Elk unfortunately has made no provisions to accommodate this.