Hacker News Reader

Comment by addicted | Hacker News Reader

Here’s what I don’t understand. I’m genuinely looking for an explanation for this.

Let’s assume everything works great for BTC. BTC is designed to eventually max out the number of coins. I believe the current expectation is that no more coins will be issued by 2040.

What happens after that? If no more coins are issued, there’s no incentive to mine. And if there is no more mining then there will be no more transactions.

Will the BTC blockchain just freeze at that point? If so why will it have any value any more?

I’m not sure what the ideal expectations are beyond this scenario.

View on HN

There are still transaction fees to encourage mining.

If you are interested in more detail, the original bitcoin paper goes in detail and is decently well written and much more straight forward to read then most people probably expect.

A lack of block rewards will incentivize all kinds of bad miner behavior:

https://economics.princeton.edu/working-papers/on-the-instab...

A nice quote:

> ... we are able to prove that an equilibrium exists. However, it is one where miners include only a fraction of available transactions into their blocks. This results in a backlog of transactions whose size grows indefinitely with time.

I agree that the game theory arguments about it all working out become much more hand wavey once rewards decline, but on the face of it, i don't neccesarily think not all transactions being processed is a problem (just retry with a higher transaction fee)

In that case, the fee will keep rising until _some_ miner puts it into a block. Mining is decentralized so empty block attacks won't work unless they all collaborate (good luck with that).

> Mining is decentralized

To me, it appears more that mining is effectively centralized in an oligopoly of mining pools, and that there's an increasing tendency towards this, compare [1] in this thread.

[1] https://news.ycombinator.com/item?id=31796847

That's because you don't understand how mining pools work

It doesn't appear so that everyone would be free to setup their self-sustained rigg and frolic on their own with realistic prospects, as the mantra of "decentralized" (as opposed to a structured environment) alleges. At least, not at a scale that would be decisive for the entire network. Meaning, there's that structured aspect and individual miners come second to this.

But can that oligopoly collude enough to actually run a 51% attack against the network?

I guess, as is, there are sufficient incentives against this, as pointed out by the linked comment. Generally speaking, when the economic pressure increases, the pressure to cooperate may increase, as well. Which may turn out both ways. (Meaning: I really don't know, and I'm not sure anyone may know for sure. Sorry for the Socratic anticlimax.) Said paper suggests that there are incentives for colluding against the system while protecting it at the same time.

The argument used to be that rising transaction fees would balance dropping mining rewards. I haven't heard that so much lately, probably because people are already not actually transacting in bitcoin due to the high transaction fees.

>probably because people are already not actually transacting in bitcoin due to the high transaction fees

That's Ethereum, not Bitcoin.

it's a bit of a chicken and egg problem. High transaction fees should attract more miners (who might be willing to accept lower transaction fees due to competition). This ought to reduce transaction fees to the market rate.

However, because there's so little real demand for bitcoin transactions from consumers, that this market rate isn't really a proper market rate. It needs to grow in volume. But the volume of bitcoin transactions is too low (by design).

So to get more bitcoin transactions to lower the rate, you first need more bitcoin transactions, but it's too expensive and consumers don't use it!

The short answer is this: The block reward is not meant to be a permanent part of the system. It is only there to bootstrap it. It's designed to slowly fade out, at which point a fully self-contained market of tx-fees will have been established, which works according to the classic rules of price-finding by demand and supply.

I imagine this will spell the end of bitcoin's (already poor) utility as a transact-able unit of value.

As others have pointed out, transaction fees approach 100% of the incentive, so fees will have to increase to incentivize the hash network to keep running. A lot of the hash power will drop out without mining on the table, and I think the net result will be that transactions will be much more expensive and slow than they are currently.

> no more coins will be issued by 2040.

The year is projected to be in the mid 2100's, so more than a full century from now. In other words, not an immediate concern.

It will be a gradual process though, with mining rewards becoming more and more rare and approaching 0 over the years.

Bitcoin issuance halves every 4 years, it's a geometric series. It's like the lily problem but in reverse.

> Lily pads are growing in a pond, and the number of lilies doubles every day. After 30 days, lilies cover the entire pond. On what day did the lilies cover half the pond?

The answer is Day 29. Very late on the timeline.

With Bitcoin, we've got halving instead of doubling, so most issuance happens very early on the timeline. In fact, 99% of Bitcoin's issuance happens within the first 28 years of the protocol's life. Bitcoin was launched in 2009. So a mere 15 years from now, in 2037, only 1% of all bitcoins will still remain to be issued.

Will 1% be sufficient to attract enough miners to properly defend the Bitcoin protocol from attack for the entire rest of its useful life? I have my doubts.

Reply on news.ycombinator.com