Will the transaction fee economy in Bitcoin follow the commission model?

1

1

I've been analyzing the long term results of the block chain. In the process I realized the relationship between the distribution of coins over time, transaction fees, and coins being lost. The full paper is available here and is about 3 pages long:

https://docs.google.com/document/d/18RirEreHjr8AVrxeL6_MeN_pQPCxWnoj1s_qPunMLX8/edit#

If this is accurate would the transaction fee system we transition to be a commission model?

Kristopher Ives

Posted 2012-08-08T08:35:47.003

Reputation: 183

Answers

1

Not entirely. In a commission model you generally pay a percentage of what the transaction is, whereas in Bitcoin you just pay for the bytes the transaction requires. The setup is less like a commission model and more like a bank transfer fee model - banks generally don`t charge you for how much you transfer, but for the transfer itself.

ThePiachu

Posted 2012-08-08T08:35:47.003

Reputation: 41 594

2That's a common misunderstanding. Transaction fees pay for two completely separate things, the marginal cost of handling the transaction and the amortized cost of hashing. The former is proportional to the data size and number of ECDSA signatures. The latter has nothing to do with those and will likely be proportional to the transaction value.Meni Rosenfeld 2012-08-08T10:41:26.303

@MeniRosenfeld, I don't think hashing cost have any influence on transaction fee, hashing difficulty gets adjusted for whatever value that would make it profitable to mine.Serith 2012-08-08T11:52:41.473

3@Serith: The difficulty needs to be high to protect the network for attackers. Transaction fees need to be high enough for mining to be profitable with a high difficulty.Meni Rosenfeld 2012-08-08T12:26:26.963

2@MeniRosenfeld: while they need to be that high, there will need to be something in the protocol which actually enables that, i.e. to avoid the much feared tragedy of the commons. I'm quite optimistic about that, though.o0'. 2012-08-08T19:02:11.090

@Lohoris The increase in difficulty is in the protocol, and is what forces everyone in the network to share the costs of the burden of security.Kristopher Ives 2012-08-09T03:14:21.700

1@Lohoris - right, I skipped that part. My bet is that there will be a hardcoded limit on the total transaction value in a Block, forcing transactions to compete on offering the most fees for a given value.Meni Rosenfeld 2012-08-09T03:53:22.837

2@KristopherIves: The difficulty will increase only if there are enough fees to incentivize miners to mine at the higher difficulty. And for that, there needs to be a way to give miners bargaining power.Meni Rosenfeld 2012-08-09T03:54:41.663

@MeniRosenfeld Handling the transaction (broadcasting it through the nodes) doesn't require a lot of resources. Moreover, the fees aren't paid to the relayers (unless you want to use Microsoft's suggestions). They are all earned by the miners that do the heavy calculations, which are a lot more costly. While the individual miners can discriminate the minimum fee based on the transaction amount, the standard client ignores it and focuses on the transaction size. Otherwise we would have a lot of addresses with small balances in inputs, as discussed somewhere else on this SE.ThePiachu 2012-08-09T09:35:11.143

@ThePiachu: Right, that's how the standard client does it now. It's not sustainable. Yes, the marginal resource cost isn't very high but it's >0 (you say yourself that it is bad if there are lots of small inputs) and should be taken into account. Currently network nodes aren't paid, but one way or another the miner's payment will eventually seep into nodes.Meni Rosenfeld 2012-08-09T10:25:35.273