I sometimes wonder if the DAYTIME protections aren't a bit of Engineering "in theory" limitations that in the real world won't make a bit of difference to protected stations. For instance, I believe it has often been stated that KABC is a no hoper because of the tower being located where it is, and the potential audience for the programming being where they are. Assuming that a power increase using the existing plant and tower is physically and simply possible, I would bet engineers could be pretty precise in what amount of power increase would be needed to allow 790 to give a city grade signal to all of Los Angeles County. I would also bet any increase would not be any daytime interference to San Diego, Fresno, Las Vegas, or northern coastal California. Since the point of this thread is what could the FCC do to 'save AM radio' why not try for an experimental period of time and allow for public comment? I bet there won't be any complaint, partly because AM listening is now so small. But such an idea could make for a viable business that could seek to put better programming on that people might listen to.
Other examples: WBWD or WSNR in New York or WFOM in Atlanta. Power increases or frequency moves should not aim to take over the world (or state), but enough to allow for a viable business.
As I suggested in another thread, AM listening doesn't need to be via a radio anymore, but it does need to have a viable footprint and 'presence' in a market to be able to attract advertisers. The AM listening will be via a well promoted internet (smartphone) App. But without a viable 'address' on the radio dial, any App is otherwise just another internet speck that advertisers will be indifferent to.
The well promoted App idea noted above could be accomplished with the FM translator as the 'address' on the local radio dial. But a lot of those translators really ARE tiny. Many times in my travels, I find the translator fades out fast, and the AM home is actually easier to listen to.