This video makes a lot of sense:
https://www.youtube.com/watch?v=1sdzoZo ... 1yXJ44maU8
Essentially the audible impacts of multipath reception are minimised by minimising the stereo image. Obviously, we don't want to broadcast in mono, but numerous tests indicate that only 10-20dB stereo separation is required by typical listeners no need for 50dB or above.
It would seem like a trivial exercise to implement in Stereo Tool? ie: instead of having a simple stereo enhancer, have something that can allow the user to select how much LESS stereo separation they want on average, but still allow boosting on tracks that have little stereo separation. Maybe have it as an option instead/as well as SSB?