You think that percentage is low. Now figure that an 1867 seated half has a survival rate of 1%. Thats low.
I don't know if I believe that. One thing is for sure, every coin show you go to seems STACKED with Morgans
Who knows for sure but have you seen many Morgans in the "wild" now a days? I would imagine if 15-17% were true, that 95-100% of that is just in the collectors realm. And if you can figure out on average how many people collect coins (and more specifically, morgans) things might make a bit more sense.
Sure, you want to take all of them off his hands for more than melt price? He'd happily sell them to you instead of the smelter. The thing is he gets a lot more coming in than he has collectors wanting to buy. The surplus ties up capital. So it gets dumped with the smelter who will take all he has. That frees up the capital for more buying. Case in point, in 1979 Unc 1883-O dollars were about $10 each and they were going begging because they were so common. Then at the end of the year silver skyrocketed and in just a couple months those $10 coins had $35 worth of silver in them. Well believe me if collectors weren't interested in buying them at $10, they definitely weren't beating down the door to buy them at $35 two months later. And Leon had 300,000 of them. Now he could have held them and sold them to collectors for $35 each and it probably would have taken years to do so. Or he could ship them to the smelter at $35 each and have a check tomorrow for the whole batch. (Not to mention that the next month silver crashed and if he had held them they would have been worth a lot less.) So he shipped them for melting. It was the correct business decision.
Also you have to keep in mind Morgans' are very heavily counterfeited coin so you have to take it with a grain of salt, any numbers you are seeing in coin shows.
I had heard, as stated by Doug, that a lot of silver coins are still melted but never really understood the economics of it. Thanks Conder, for taking the time to provide a great explanation. As for the OP, setting aside proofs, I would think a 15% survival rate for any series from the 1800s-early 1900s is actually quite high. Silver dollars are not aslikely to be lost as coins of lower denomination, and other than in some parts of the West, were not as actively used in commerce.