There doesn’t seem to be clarity on what exactly a “hash war” looks like. Some people reject the possibility a hash war is even possible when replay protection is used.
“They vote with their CPU power, expressing their acceptance of valid blocks by working on extending them and rejecting invalid blocks by refusing to work on them. Any needed rules and incentives can be enforced with this consensus mechanism.”
– Satoshi Nakamoto, Bitcoin Whitepaper
The idea of voting with CPU power, as explained in the quote above, does not seem to work when replay protection is added. Normally, if miners disagree, the chain with the most proof of work is considered the “valid chain”.
Replay protection, however, allows two chains to use the same proof of work algorithm with different rule sets for which transactions are valid. For instance, when Bitcoin Cash forked from Bitcoin they used replay protection so that transactions on BTC would not be valid on BCH. Blocks for each chain were different thus guaranteeing two separate chains, creating a peaceful split in rules.
How can a chain split be prevented when replay protection is added? Could BTC supporters have prevented BCH from ever existing? A chain split cannot be prevented simply by accepting valid blocks and rejecting invalid blocks. However, if a majority of hash power wants to actively attack a minority chain, is it possible to make it economically infeasible for the minority hash power to protect it? Surprisingly, the economics give the majority hash rate a large advantage over the minority in a hostile attack - even if market prices favor the minority chain.
Hypothetically, let’s say in the current Bitcoin Cash conflict that SV was able to get 60% hash power support and ABC had 40% hash power. Let’s also say ABC coin is trading on exchanges for $500 while SV coin is only trading for $100.
| Bitcoin SV | Bitcoin ABC | |
|---|---|---|
| Hash rate | 60% | 40% |
| Market price | $100 | $500 |
In our hypothetical, when November 15th hits and the last common block is mined, initially SV supporters would mine an SV block and ABC supporters would mine an ABC block. However, since SV miners decided to actively attack ABC chain, they would start mining empty blocks off the last common block (attempting to orphan the first ABC blocks).
Jump forward to after 100 blocks have been mined total. Since ABC is trying to be peaceful, their 40% hash rate would have mined 40 ABC blocks. SV supporters on the other hand would have attacked ABC with empty blocks. Their 60% hash rate would be able to mine 41 empty blocks, orphaning all ABC blocks mined by ABC supporters. They could also mine 19 SV blocks, keeping their chain alive.
| Bitcoin SV | Bitcoin ABC | |
|---|---|---|
| SV blocks | 19 | 0 |
| ABC blocks | 41 (all empty) | 40 (orphaned) |
What would the profitability look like here? Let’s assume fees are negligible and only look at coinbase transactions. Each new block generates 12.5 new bitcoin for the miner - at $500 USD per BCH that’s a little over $6,000 per block of revenue. For simplicity, let’s assume it also costs on average $5,000 to mine a block.
With these assumptions, the 60 blocks SV supporters would have mined would cost $300,000 and the 40 blocks ABC supporters would have mined cost $200,000. However, on the revenue side, the 19 blocks SV supporters would mine for SV chain would bring in $23,750 and the 41 ABC blocks would bring in $256,250 for a total of $280,000. Since all ABC supporter blocks were orphaned, they would have no revenue.
Looking at overall income, SV supporters would spend $300,000 and earn $280,000 for a loss of $20,000. ABC supporters would spend $200,000 but earn nothing, for a loss of $200,000.
| Bitcoin SV | Bitcoin ABC | |
|---|---|---|
| Cost | $300,000 | $200,000 |
| Revenue | $280,000 | $0 |
| Income | -$20,000 | -$200,000 |
What happens next? Would ABC supporters continue to mine blocks that keep getting orphaned with no profit? Would the market continue to support a chain that does not confirm any transactions? My guess is no to both. Miners would stop mining ABC chain and the market would stop valuing it at $500.
The next question is whether the market and/or the miners would move to SV. Considering how much people have invested - not just buying coins, but creating businesses, etc - I don’t think the market would abandon the only remaining BCH chain. At least some percentage of ABC users would begrudgingly use SV if ABC was not a choice.
If things were to play out like this, it would mean, for better or worse, that miners have ultimate power in bitcoin. It would lay the road map for settling future disagreements with hash power. It could even be a trial run for a stand-off between BCH and BTC.