I have a few comments to elaborate on my own opinions on this.
May it goes without saying, but this is first and foremost. A race with 20 participants doesn't succeed, and it doesn't matter if those 20 are fast or slow. The fact that a race has 40,000 people isn't the problem. The fact that a race has 40,000 people and doesn't accommodate those that feel it's a race is. That doesn't mean making sure the 12 invited running are in the front. That means some reasonable seeding so those that feel it's important have an opportunity to be seeded and those that don't or aren't ambitious enough to read a brochure or web page end up in the back.
I agree to a point, but I look at it somewhat differently. Sure, there are parts of the country (like mine) where a Kenyan B or C team will show up because there's money. The local sub-elite could compete for the most part, but if you're the only money race that month you'll get all of the foreign road whores. Like it or not, the local Chevy dealer and the Chamber of Commerce want to see someone with local ties on the podium. Nothing wrong with road whoring in my book, but we need to get more opportunities for some whoring by folks that have some regional ties.
and if an old gimp like me wants to pay $100 to run a marathon then why not? If it's worth it to me then fine. Supply and demand and all that. If someone can run the top 1% of times then they should get in anywhere regardless. Not comped, but guaranteed entry. If you're a 2:40 guy well that's fan-f**kin-tastic but the fact is you're not a factor in a major marathon, nobody is going to come to the expo because you're there, and if you're coming in from out of town no local's give a s**t anyway. If you're the home town hero, any race director worth their salt gets you in.
So the brass tacks is I don't believe the trails qualifier time is killing USA (or helping it), same way I look as Boston qualifying times. I'd recommend lowering the Boston times because the vast majority of the people that filled next year's race QUALIFIED. Even if the top end projections of 30% charity runner are true the fact is 70% made the standard. Do I think that will make marathoners over the nations faster? Not a bit. I'm just saying what I would do to manage demand.
I deal with red herrings every day. With 1000 some odd developers worldwide every time there's a performance problem I hear "engage the network team to look at the firewall/gateway/load balancer". It doesn't matter than there isn't one of them in the equation, and that the physical difference between what they are talking to and from is measured in inches. The person raising the issue didn't do their homework and wants to close out the call by passing the issue on to someone else.
The key is knowing context. If a developer in Wuhan is testing from their laptop and going through 14 firewalls and god knows how many switches and failing performance tests, then we ask why the tests failed. If the conditions for the test turn out to be unrealistic we work to make it possible for them to run those tests locally to achieve equivalent results. Do we say they're a wuss because the laws pf physics conspired against them? No we look at the core reason, and work up a way to correct it.
I look at US distance running the same way. I see people saying the problem is Runner's World, long race cutoff times, Boston qual time, trials qual times, charities, etc. All whipping boys, but frankly all red herrings. Let's face it. There may have been a s**tload of sub 2:20's in 1983 at NYC but there still wasn't an American on the podium at the Olympics in 1984, no American broke 2:13.