So the point of reward halving (among other things) was to attract users to this then entirely new technology. I take it, and it might have served well to this end, but I'm still suspicious that nowadays it wouldn't make more harm than good...
It is in fact a large topic about reward halving, I guess Satoshi must have put lots of consideration in this area
There are basically 3 types of money supply scheme:
1. exponentially increasing (fiat money)
2. linearly increasing (Dogecoin)
3. limited (Bitcoin)
In fact, 2 and 3 are almost the same from macro economy point of view: After some decades, the increase in total money supply in 2 will become so insignificant percentage wise, that it is almost like fixed (If the money supply increases 0.1% per year, then its impact on inflation is almost the same as limited total money supply)
So for the sake of simplicity, Satoshi went for model 3, which could give a very clear signal of limited total money supply
Then he had to decide how those limited coins will enter circulation over time
There are basically 3 types of distribution model:
1. distribute using a progressively increasing schedule
2. distribute equal amount of coins each year for xxx years until reach the limit
3. distribute using a progressively decreasing schedule
It is clear he didn't want to use model 1 or 2: If bitcoin indeed take off and gained mainstream popularity, then large banks with lots of resource will easily control majority of the coins at later stage. Those who are interested in the idea must be able to control a large part of the bitcoin when it is still in its infancy
So a progressively decreasing schedule was selected. Then how fast and how often the block reward will drop will be decided. It must consider the speed the information is spreading on internet, give early pioneers enough time to react, but it should also leave enough incentive for late adopters to continue strengthen the network until the fee income replace the block reward
If the reward drops too fast, it will reduce the late adopters' incentive thus might fail to attract enough support before it reach mainstream. If the reward drops too slow, it will be close to model 2 thus give large institutions enough time to control majority of the bitcoins. 4 years period is already a long time for any IT related projets, so let half of the coins distributed during 4 years sounds like a decent pace
Why 50% drop every 4 years instead of 16% every year? I have not figured out a good reason, maybe to make the mining anticipation simple and relatively stable during a 4 years time. Could also be a strong anticipation of price rally every 4 years, thus everyone is at least hold their coins for 4 years, encouraging long term saving
