Why not decrease the difficulty to increase capacity?

Question

There is a persistent debate in Bitcoin about increasing the capacity.

The popular idea seems to be to increase the blocksize from its current 1MB to push through more transactions per block. But this also risks forking the chain.

I think a much simpler idea would be to reduce the target time from 10 minutes to 5 minutes. This would make hashes less difficult (fewer leading zeros), and basically target 2 blocks per 10 minutes instead of just 1.

I'm confident there is some downside to this idea, or it would have caught on already. But I'm just not seeing it.

What is wrong with decreasing the target difficulty to increase Bitcoin's capacity?

Answer 1

One problem is that lower block times mean an increased chance of forking, which makes head of the chain, and the system, less reliable.

Another problem is that it normally takes a block a few seconds to a minute to propagate across the network. Proportionally that time becomes a lot higher with a decreased block time, giving the original miner and miners with direct connections to them a more significant lead time.

Finally, it throws off the whole schedule of when Bitcoin will be released. New Bitcoins are set to be mined past the year 2100, that would be significantly sped up, changing a lot of existing assumptions about inflation and economic assumptions about scarcity of the currency. Since the change would have to be a hard fork it would be controversial both economically and technically.

Answer 2

Steve Ellis' answer illuminates the theory behind the initial choice of smaller, less frequent blocks well.

In practice...

Since your question was about "reducing" the difficulty, meaning you were referring to a fork that would change the rules of Bitcoin, a far bigger tradeoff would be that a change to consensus like this would ultimately have to be introduced by a group of people and agreed upon by enough of the network to force a chain-split. A complete chain split has many considerations that is not really a technical question, but rather a debate on politics, sociology, economics, and game theory. Essentially, everyone (miners, nodes, holders) would have to agree that the hard-forked coin is Bitcoin.

The inability of anyone or any group to change these rules is crucial to the value proposition of Bitcoin at this point, and the damage it would cause to trust in the system as an immutable one would be far, far greater than the slight loss to decentralization that the faster block times would bring.

Reducing the difficulty like this would brick ASICs, so you're not going to get the miners, who have spent hundreds of millions on equipment already.

This challenges the immutability of Bitcoin, so you're not going to get the nodes or the holders.