Is it any wonder that the generations who still have most of their hearing have largely abandoned over-the-air radio? They may not be able to quantify why they dislike it, but at some level they understand that FM generally sounds like crap and AM sounds like an old telephone landline.
Looking back, we can see that the period 1970-2000, más o menos, was the exception to the general rule of radio sounding obtrusively unnatural. That's when FM took over from AM and, while some FM stations could torture their audio to sound LOUD and fatiguing, many others took a reasonable approach to processing, sacrificing some dynamic range but still managing to sound fairly natural. Before 1970, AM radio predominated with FM being a niche market. You put up with AM because that's what you had. In the 1970s, FM became increasingly mass-appeal; in the 1980s and beyond, FM was dominant. Around 2000 with Napster, et al., people started listening to music through data compression algorithms that had a
different sort of unnatural sound, at least to some people. The purpose of going back through all this is to point out that you and I were socialized to the sound of FM radio. It wasn't always perfect but it could be pretty good in the right hands. Then file-sharing and streaming came along, socializing subsequent generations to sound that's been filtered through data compression algorithms. The generations that once put up with some of the defects of FM radio - hiss, occasional static in stereo, multipath distortion - have made way for generations who have traded off those defects for the defects of audio data compression algorithms. As bandwidth becomes ever more available, look for the next frontier in streaming to be lossless algorithms...for a price, no doubt.
Of course, the other major defect of FM radio is programming. People came to FM for the music and tolerated commercials as the price. Post-2000-ish, they don't have to do that. It's a structural defect. Format labels and implementation trivia no longer matter. Radio's survival will depend upon finding new business models and approaches. Program directors who are stuck in repeating what worked in 1985 or 1995 are probably not going to have the answer to that challenge.