Op-Ed: Bitcoin’s High Fees are Forcing Companies to Optimize Their Interactions with the Blockchain, And That’s a Good Thing

0
2710

There have been quite a few proposals to raise Bitcoin’s block size limit by way of a hard fork over the past three years, but none of them were able to gain consensus among the nodes on the network. The key selling point of such an increase to the block size limit has always been the idea that an increase in the supply of block space will lead to lower transaction fees and the ability for the network to process more transactions per second, but the current congestion seen on the Bitcoin network has actually had a positive impact in terms of the forced optimization of companies’ interactions with the blockchain.

While higher transaction fees have undoubtedly priced out some low-value use cases of Bitcoin block space, the optimizations being implemented by bitcoin custodians and wallet providers should help the cryptoasset network retain its core value proposition of decentralized, censorship-resistant digital money over the long term.

Throwing Trash Into the Blockchain

In the early days of Bitcoin, no one really cared what sort of data was being tossed onto the blockchain. While there was technically a hard block size limit set at 1 MB, this effective limit on the availability of block space was not taken seriously because the idea of full blocks was a far-off concept at that point.

As an example of the thoughtlessness in terms of block space conservation from the time, the online gambling platform Satoshi Dice placed every single one of their users’ wagers on the blockchain and at one time accounted for roughly half of all transactions processed by bitcoin miners per day.

While some have referred to this sort of activity as spam, others have claimed it may be more accurate to call it “blockchain pollution”.

The Satoshi Dice example is just one of many in terms of companies not caring about optimizing their interactions with the blockchain because block space was plentiful at the time.

A desire for more space to handle this sort of blockchain pollution may have been a key, driving force behind various Bitcoin companies’ support for hard-forking increases to the block size limit via Bitcoin XT, Bitcoin Classic, Bitcoin Unlimited, SegWit2x, and now the Bitcoin Cash (Bcash) altcoin. This sort of hard-forking increase would effectively kick the can down the road in terms of optimizing the use of a scarce resource (block space).

It’s unclear how far down the road the can would be kicked before entities began to optimize their use of the blockchain in a situation where the block size limit had been raised through a successful hard fork, as there would be practically no incentive — in the form of costly fees — to implement those optimizations.

Bitcoin Companies Become Blockchain Conservationists

Now that the median on-chain Bitcoin transaction fee is around $15, bitcoin custodians and wallet providers are suddenly much more interested in optimizing their use of the blockchain.

It started with transaction fee estimations. Pretty much every bitcoin wallet was not prepared to handle a fee market caused by full blocks, which meant users were left with transactions that would practically never confirm on the network once it became more congested in early 2016. Transactions fees became less predictable at this time, and most wallets have no way of altering the fee attached to a transaction based on the current supply and demand around block space.

A proper market for fee estimation is critical to the long-term success of Bitcoin, as transaction fees are what will incentivize miners to secure the network as the block subsidy slowly disappears.

Fee estimation in wallets has definitely improved since early 2016 out of necessity, but there is still more work to be done here.

Another change incentivized by high fees has to do with avoiding user error more generally. If fees are high, then companies will not want their users to be making more on-transactions than what is absolutely necessary.

Recently, BitPay implemented the Payment Protocol, which is intended to reduce the number of extra, costly on-chain transactions caused by user error. BitPay explicitly pointed to high transaction fees as their reasoning behind their Payment Protocol integration in a recent blog post.

Transaction batching and Segregated Witness (SegWit) are two other changes that are now becoming critical for exchanges and other types of bitcoin custodians to implement.

With batching, custodians are able to combine a large number of payments into one on-chain transaction on behalf of their users. Bitcoin researcher and writer David Harding has estimated that transaction fee savings of up to 80% are available through batching.

With SegWit, Bitcoin users are able to gain access to more block space (enabled through a soft fork), which leads to an effective increase in transaction capacity. According to BitGo, their users are able to enjoy a 50% reduction in transaction costs thanks to SegWit. It should be noted that the benefits of SegWit in terms of transaction costs are more impactful in a multisig-focused wallet such as BitGo.

 

Blockchain and Coinbase, who are two of the biggest creators of on-chain Bitcoin transactions, are currently implementing SegWit. Blockchain is unable to implement batching due to the fact that it is a non-custodial wallet, but Coinbase is working on the batching optimization as well.

Other platforms, such as Bitstamp, have already implemented batching and SegWit.

Again, it’s unclear when these companies would have been incentivized to optimize their interactions with the blockchain if fees were not allowed to rise.

These sorts of optimizations enable Bitcoin to retain the core property of permissionless access because the costs of operating a full node are able to remain relatively low when the blockchain is used more efficiently.

On other cryptoasset networks, such as Bcash, the low fees may lead to more blockchain pollution and a model that does not incentivize more efficient use of the blockchain. This may lead to more centralization and a loss of the blockchain’s key value proposition over time. To this point, the aforementioned Satoshi Dice now operates on the Bcash network.

Further Improvements Possible

Of course, there are also other options either available now or coming soon for bitcoin custodians and wallet providers to make more efficient use of the blockchain.

Specifically, there isn’t much of a good reason for bitcoin custodians to broadcast every transaction between their users on the public blockchain. Instead, the custodians could keep track of the transactions between their respective users off chain and then settle on chain once per day.

Moonbeam and Liquid are two projects that would allow custodians to transact with each other in a low-trust manner without having to touch the blockchain on a regular basis.

Of course, the Lightning Network is sort of a more decentralized version of Moonbeam and should also enable this sort of activity between custodians.

Getting back to the point of transaction fees creating incentives to optimize interactions with the blockchain, longtime Bitcoin Core contributor Matt Corallo sketched out a plan for a hub-and-spoke model of payments channels (the same technology used in the Lightning Network) as a method to scale Bitcoin and sent it over to Blockchain in early 2014. According to Corallo, this plan was ignored.

“Ultimately, I get it, Bitcoin businesses have always been fighting to keep afloat of technology in this space, so telling them ‘hey, here’s this awesome technology that I’d love to work with y’all to finish implementing, because I think it’ll make Bitcoin scale’ when they aren’t facing scaling issues at that time just isn’t a compelling argument,” Corallo commented on Reddit.

Now that these Bitcoin businesses are facing issues in terms of network congestion and higher transaction fees, the incentives to better optimize their backends are present.