The Bitcoin blockchain is basically becoming useless for low-value, instant purchases. Frankly, if you're ordering a beer at a bar, you're better off using cash or a credit card at this point.
Bitcoin was once touted as a cheaper alternative to credit cards that was going to revolutionize the payments industry, but the P2P digital cash system has hit a bit of a speed bump in terms of those long-promised "free" transactions. Let's take a closer look at Bitcoin's increased transaction fees and what solutions exist to solve this issue.
What's Wrong With Small Payments on Bitcoin?
The problem with low-value transactions on the Bitcoin network right now is that fees are too high for it to be of any practical use in these situations. 21's bitcoin fee estimator recommends a fee of 54,240 satoshis at the time of this writing, which equates to roughly $0.69. This cost is roughly double the amount of fees involved in many low-value credit card transactions.
In fact, as a user on Reddit pointed out , more than half of the Bitcoin addresses in existence contain a balance lower than the current estimated transaction fee . However, the balances of these addresses amount to roughly 0.00001% of all bitcoins in existence.
While the level of censorship resistance offered by Bitcoin's permissionless payment network has obvious value, it's also rather expensive. The system must remain decentralized in order for this resistance to censorship to remain intact. The cost of operating a full node increases as more transactions are processed on the network, and some users already see the costs related to full node operation as too high .
The reason that this sort of change has not yet been implemented is that coming to consensus on a proper solution for this sort of issue in a decentralized fashion has proven to be difficult. It is both a positive and a negative that Bitcoin is resistant to these sorts of changes.
Bitcoin payments were much cheaper in the past because demand for block space was much lower. Users are now essentially bidding on the right to have their transactions included in the next block. Large-value transfers and payments that necessitate a certain degree of censorship resistance will tend to outbid lower-value, everyday payments such as the purchase of a cup of coffee.
How Could Smaller Payments Work on Bitcoin Right Now?
For now, smaller-value payments with bitcoin tend to require a degree of trust in another party, which is ironic because removing that third party was outlined in the original Bitcoin white paper as a reason for Satoshi Nakamoto 's creation. If you're going to avoid an on-chain transaction fee with your bitcoin payments, then you'll likely be holding your coins in a bitcoin bank, such as Coinbase or Xapo , with the receiver of your payment also having an account at that particular bitcoin bank.
The level of trust in the bitcoin bank can already be mitigated a bit by using something like Blockstream 's Liquid or an Open Transactions voting pool . In these sorts of schemes, funds are held in one giant multisignature address by all of the banks involved in the program. An individual bitcoin bank would not be able to steal customer funds, but customers must still trust that a majority of banks (depending on the setup) do not collude to steal funds from the now much-bigger pot.
Systems like Liquid and voting pools also have the side effect of not requiring an on-chain transaction for the transfer of funds between the participating entities and their customers.
No bitcoin bank or exchange has publicly announced that they are using one of these systems at this time.
Alternatively, bitcoin banks could decide to make agreements with each other to settle transactions between their customers once per day. This would essentially mean that payments between customers at Coinbase, Xapo and any other centralized bitcoin bank or exchange would happen off-chain at little or no cost.
How Could These Systems Improve in the Future?
In the spirit of Bitcoin, the level of trust required in these centralized entities can also be reduced through further improvements at the base protocol level. Solutions such as the lightning network and TumbleBit drastically reduce the amount of trust required in one of these centralized institutions to the point where it would probably be incorrect to refer to them as banks at all.
In TumbleBit, for example, the "bank" is unable to steal customer funds and does not know where user funds are being sent. TumbleBit also provides better privacy and essentially instant quasi-settlement of bitcoin transactions.
Less useful and elegant versions of the lightning network and TumbleBit are possible with the current version of Bitcoin, but they would be greatly enhanced if Segregated Witness were activated on the Bitcoin network. Content sharing and monetization platform Yours intends to launch with a payment solution similar to the lightning network that acts as more of a hub-and-spoke system rather than a generalized layer for these types of payments.
In a situation where these layer-two protocols are implemented and widely used, bitcoin users would essentially have two sets of bitcoins in their wallets that come with different levels of security. The bitcoins that are unconnected to the lightning network or a TumbleBit hub would be like money in a savings account, while the coins "tied up" (for lack of a better term) on these layer-two systems would be similar to money in a checking account.
Then again, this structure would not be too dissimilar to how things work today. Many bitcoin users already keep their bitcoin savings in some form of cold storage while their spending money is in a mobile wallet. The specifics of how this would play out are somewhat unclear at this time.
An increase in the block size limit via a hard fork would also create an increase in the supply of block space, and thus lower the cost of on-chain transactions overall, but it's unclear how large blocks can become while retaining a sufficient level of decentralization and censorship resistance. A key difference with the aforementioned solutions is that they potentially increase the level of risk involved with a user's funds on an opt-in basis. With hard-forking changes being made at the protocol level, users are forced to opt into new risks by default.
This is not to say that a hard-forking increase to Bitcoin's block size limit cannot be made; it just needs to be balanced with the tradeoffs made when it comes to the cost of operating a full node. Past research has indicated that an increase from 1 MB to 4 MB may be worth it, but new research in this area would be useful.