Why a Bitcoin hard fork can't and won't stop the original Bitcoin chain

published Aug 04, 2016

No matter how you try, you can't force your blockchain on anyone else. We have a good idea of how such an attempt would turn out, so let's go over it.

Why a Bitcoin hard fork can't and won't stop the original Bitcoin chain

Some people who would really like to see Bitcoin hard-fork (to accommodate their pet ideas) believe this:

If the [Bitcoin] Core blockhead [sic] devs were to only agree, then the whole system could easily "fork" [to a chain with bigger blocks] in such a way that there's only one viable chain left.

We can infer from this opinion that "success" w.r.t. a hard fork means that a vast majority of people will (be forced) to move to the challenger chain, to such an extent that the original chain stops working.  That is:

  • a) there will be a hard fork, which will produce two chains and two cryptocurrencies:
    • the original chain and currency, with chain parameters they do not want;
    • the challenger chain and currency, with chain parameters they want (bigger block size, et cetera);
  • b) shortly after the hard fork, the challenger chain will take over the original chain, with all economic activity switching to the challenger chain;
  • c) the original chain will become nonfunctional or abandoned.

Let's take these words at face value, and stipulate that this outcome means "success" in the context of this post.  Let's furthermore stipulate (a) and (b) — in other words, that pro-hard-fork people actually get a hard fork and a challenger chain running, which is successful enough to attract attention from many Bitcoin actors.

So, is there any chance of outcome (c) happening?

I'll start with the conclusion: unfortunately, people who want that outcome will discover — much to their chagrin — the original chain cannot and will not stop running.  It is quite literally impossible to stop it, and already-existing economic examples show us why it will not happen.

Now I'll explain why.

Can the creators of forking Bitcoin software render the original chain unusable?

Let's first address the question of how "a fork can be made so that the original chain can't run anymore".

No one can stamp out existing open source software. This option cannot be taken away. Additionally, any improvements on the challenger fork codebase can be easily merged into a copy of the original codebase, which guarantees there will be up-to-date and functioning versions of the software that respect the original fork's parameters.

Nothing short of automatic and pervasive online censorship can stop that from happening.

This means that people — users, sellers, holders, buyers, miners, traders — will have the option to continue running on the existing chain.

Would people remain loyal to the original chain?

As for whether people would continue running that existing chain, the answer is yes, some would.

Why? Profit.

As miners move to the challenger chain, the original chain miners collect more fees and a larger share of the block reward.  So, the profitability of the original chain rises over time during that exodus.  At some point in that exodus, the profitability of mining on the original chain will be high enough that some miners who went for the challenger chain will decide to come back.  Like a (very slow) guitar string, miners will switch back and forth until profitability reaches a (complex) equilibrium.

We know this is the case because we already saw exactly that happen with the people who mine the most profitable Ethereum chain at any point in time.  There is widely-available and cost-free software which can tell you which chain was more profitable at any instant in time — and Ethereum miners use the output of the software to decide which chain to mine on.

As we know from experience, contrary to the belief that "once the majority leaves the original chain everyone else will leave it too", what ends up happening is that hash power between the two chains will be stabilized by miners swapping swap back and forth, as they attempt to collect the maximum profit from their investments.

We can therefore conclude that the original chain will continue to be maintained by miners.  Whoever chooses to continue to run miners and validators, will maintain the original chain running and operating normally (if at a temporarily reduced hash rate).

Furthermore, users will continue to trade coins on one chain vs. coins on another chain, seeking their own profit motive.  Again, contrary to the assumption that the market prices of coins in the original chain will plummet to zero, we'll see the prices of coins in both chains bounce up and down (with trends in opposition to each other) until they find a complex equilibrium.  We know this because we saw that very thing on the Ethereum side — the original Ethereum tokens plummeted to near zero, then they went back up again.

This profit opportunity will inevitably prompt people who take coins to accept coins on either chain.

Nothing short of automatic and pervasive Internet traffic sabotage belonging to a specific chain can stop that from happening.

How does a post-hard-fork world look like?

The end result of such a hard fork will be as predictable as the existing ETC / ETH fork.

  • Two chains, not one.
  • Combined cap of the two associated currencies might or might not be higher than the previous original chain, but each chain's cap will be less than before — the price of coins on each chain will probably go down, possibly halve.
  • Interestingly, people who held Bitcoin on the original chain will not technically lose purchasing power, as they will be able to spend them twice — once as the original currency on the original chain, and once as the new currency on the new chain. Some will sell original coins to buy challenger coins. Some will do the opposite.  Arbitrage between the two chains will be normal.
  • Ultimately, this otherwise-superfluous economic activity will hash itself out, as we settle into the reality of two permanent chains.

That is the way things will go. There's nothing the hard forkers, or any developer, no matter what code they control, can do about it.

Now you know why, ultimately, a block size fork will simply be a make-busy-work operation. Economics and human action will it. People who deny this outcome simply do not understand, systemically, how the software works, what the motivations are of the people running the software, and how the economics of such a change work out.