#2: Don't agree with. Scarcity is good. Even if you can break it down to 1s, I think the $21MM coins max module.
#3: Semi agree. Problem with this is that then we need to have $356 Billion coins in circulation like Mooncoins.
Which are useless btw.
Regarding coin supply, consider that Satoshi could have launched Bitcoin with a grand total of 2.1 BTC as the maximum number of base units rather than 21 million BTC. The way this would work is that instead of mining 50 coins per block you would mine .00000500 per block. Likewise, he could have launched it with 21 quadrillion base units (Satoshis). You see how these are just numbers? There is no difference between the large and small numbers here as long as the ratio is the same. If the ratio of mine-able units to the total base units (or whole units, or whatever you want to call each "coin") is the same, then these wildly different looking releases are actually the same. The only difference is the perception.
So we see that the number of base units issued is quite arbitrary and meaningless in itself. What matters is the public perception of scarcity and how that affects adoption/speculation, and also the ratio of mine-able units to base units. Keep in mind that in any case of a crypto release like Bitcoin, there is no question of ever having a hard limit on trade-able units since all it takes to inflate the supply of trade-able units is to add another decimal place in the software when needed! This need would only arise after the value had risen considerably and the smaller divisions would be too valuable to be of use for the smallest purchases.
So what does the coin supply actually mean then? I think by this analysis it means that we are simply looking for the proper way to stimulate buying and saving/hoarding over time while also trying to have the greatest adoption rate and the most even distribution possible without sacrificing one of these factors for another. And of course, the ratio of mine-able units to base units must be great enough to ensure profitable mining for as long as possible too - and hopefully, again, a very wide base of mining.
You laugh at Mooncoin, but what's the real difference? Issuing 356 billion
might have been better for Bitcoin. The number itself is only relevant to perception and how that affects our behavior, but the effect might have been a much larger base of users and a slower rate of price increase over time. By going with 21 million, Satoshi ensured a rapid rise in price if it caught on due to the heightened perception of scarcity, but did he sacrifice a very wide base of users and a more modest stable price graph? What if Bitcoin was able to hold a $1-2 price up until now? How would that affect adoption? Would Grandma and little Billy and cousin Jep all be jumping into Bitcoin now instead of the relatively few number of enthusiasts and speculators? What kills is that we don't know the answer to this question. It's really tough to say, but it seems like there's good reason to believe it would have been better to cool this thing down and spread it deep and wide so that it would be more accessible to the average person over a longer period of time.
Who says cousin Jep and Grandma aren't in Bitcoin right now?
But seriously, I totally understand you're point. But I don't think it matters. Look through the actual trades of most bitcoin trades, they aren't 5 coins, 2 coins. They are very small fractions, .002, .04, etc. Small investors mainly, and maybe.
Looks at the stock market also. If a penny stock has 1 Billion shares outstanding, it almost no chance of becoming anything significant. On the other hand, look at something like Telsa or Netflix. Both relatively small float, so when positive news comes the stock shoots up significantly. Same thing happened to Bitcoin. And bitcoin will never "split". Berkshire will never. Google said they won't either. I think that's a good thing. Will we ever catch up to BRK?! Probably not anytime soon, but you never know!