Quote from: ribuck on May 09, 2011, 09:43:48 AM

On the day before the reward decreases from 50 to 25, there will be approx 7200 new coins generated out of a total of 10.5 million. Less than 0.08% extra. On the following day there will be approx 3600 new coins generated out of a total of 10.5 million. Less than 0.04% extra.

The change from 0.08% to 0.04% per day won't upset the markets, and the miners will have known about the coming change for ages, and will have factored it into their plans and budgets.

This is not something to sweat about.

Halving the reward is the equivalent of having half the gold content in your ore disappearing over night. This will cause most miners to become unprofitable, so block generation could crawl to almost a halt, making the next 2016 blocks very long and painful.

I think the 2016 block interval for difficulty changes could cause big problems, not limited to times of reward change. Suppose the market tanks at the beginning of a new difficulty, causing 90% of the hashing power to quit - this would mean that it would take 20 weeks until difficulty was adjusted again, not good...

This is unless fees go up a lot or are already a major part of the reward for each block.

Is this correct reasoning?

Just because you see an event happening does not mean it won't cause major changes in a market if you are at a loss at what to do about it...
"Tomorrow God comes down and removes half the gold in your ore" is a smaller problem, since the security of transacting existing gold does not depend on miners to keep mining. Gold miners will resume when/if the gold price makes it profitable.

So I think there clearly is a problem. The solution to this problem should be pretty obvious and, I believe, technically easy to implement. It also should not be that big of a problem politically, but I would like first to see if people agree it is a problem, and comes up with the same solution.