A Brief History of EIP1559 for Busy People: No, It Won't Guarantee an Increase in Ether

This article is authorized and reproduced from Taipei Ethereum Meetup, written by Ping Chen. The original title is "A Brief History of EIP1559 for Busy People"

1. Problems to be Solved

Competition in Transaction Fees

Currently, blockchain networks including Bitcoin and Ethereum have effective limitations. Bitcoin's TPS (transactions per second) is 7, Ethereum is around 15. When the number of pending transactions exceeds the blockchain's processing capacity, the validators (miners) responsible for producing blocks will select transactions with higher fees to include in the block, letting the market mechanism determine the order of transaction processing.

While this design seems reasonable at first glance, it adds an extra cognitive burden on everyday users.

Ordinarily, when we send a transaction, we hope it will be processed quickly but at a lower cost. This leads to a strategy of "checking what others are offering" before slightly outbidding them. This requires users to "predict" the congestion level of the blockchain in order to squeeze in the next block with the least amount of money, adding to their troubles. Moreover, when many people are eager to send transactions, fees will be ruthlessly driven up until most people can no longer afford them. These excessively high fees end up in the miners' pockets, which is not the most optimal in terms of fee distribution.

Lack of Flexibility in Block Resources

Another problem that EIP1559 aims to address is the peak-hour scheduling of blockchain resources.

Currently, each block has a fixed transaction capacity, but demand fluctuates. Typically, there are more transactions on weekdays than weekends, and evenings to late nights in Asian time zones see more activity than during the day. Occasionally, sudden demands like ICOs or NFT sales can dramatically increase fees to an extreme level, temporarily paralyzing the blockchain for users unwilling to pay such high prices.

2. Evolution of Solutions

Second-Price Auction

Originally, transaction fees were based on how much you bid (first-price auction), leading users to carefully choose a not too high or too low amount. With the second-price auction approach, "regardless of the initial bid, transactions within the same block are charged the same fee," determined by the lowest bidder in that batch of transactions. This way, users don't need to overthink, just bid the highest price they are willing to pay, as any excess over the lowest price will be refunded.

Burnt Transaction Fees

However, the second-price auction has a significant flaw: it can be manipulated by miners. Once miners arrange transactions from highest to lowest, they can replace the lowest-priced transactions, intentionally creating some high-fee useless transactions. Since the fees ultimately return to the miners, inflating the lowest fee also forces the transactions ahead to pay more to the miners, increasing the miners' profits.

https://vitalik.ca/files/misc_files/EIP_1559_Fee_Structure.pdf

To address the issue of fee manipulation by miners, the simplest solution is to burn all transaction fees! Miners using their own money to inflate fees will only lead to more losses.

Base Fee + Tip

EIP1559 ultimately settled on an intriguing solution: the system automatically adjusts fees based on demand.

First, the original block size limit is doubled to the target. If a block is expected to use 15,000,000 gas, the limit is set to 30,000,000 gas. Miners can still fill the block, but the fullness of the block determines the base fee for the next block, with a 12.5% fee adjustment for each block.

For example, if the base fee is originally 20 Gwei and the block is half full with 15M gas, the base fee for the next block remains at 20 Gwei. If the block is empty, the fee drops to 17.5 Gwei for the next block; if the block is filled with 30M gas, the fee increases to 22.5 Gwei for the next block.

The design of the new system is similar to the market supply-demand determination of the second-price auction but significantly reduces the miners' manipulation space. It provides more flexibility for the blockchain to handle sudden high demands, allowing the system to temporarily process transactions at double speed. Although this will quickly raise fees, they will gradually decrease during off-peak hours, essentially borrowing some capacity from the future.

However, in situations like the sudden congestion caused by NFT sales, the double capacity may still not be sufficient, and the 12.5% fee increase per block may not be enough to dampen buyers' enthusiasm. Thus, EIP1559 retains the "tip" mechanism, allowing users to pay miners to accelerate transactions in extreme situations.

3. Community Reactions

Angry Miners

This development may have been unexpected for developers.

Disputes between Ethereum core developers and miners are not new, including previous reductions in block rewards and the cancellation of the ProgPoW upgrade, which have led miners to threaten actions. Moreover, in the visible future, PoS will render miners completely unemployed. In comparison, this time it is just about removing fee income, and miners should theoretically have already become accustomed to the changes.

However, the DeFi liquidity mining boom in 2020 led to high transaction demands, with fees often exceeding hundreds of Gwei for extended periods, shifting the miners' fee income structure from 5-10% to almost a 1:1 ratio with block rewards, or even exceeding them. In this context, the proposal to burn fee income naturally provoked strong reactions from miners.

Collapsing miners began various complaints on social media, accusing developers of taking their money and claiming that this change would compromise blockchain security (in reality, high fee proportions can lead to blockchain reorganization). Blockchain columnists friendly to mining pools also criticized and accused the proponents of EIP1559 of economic fallacies, suggesting a complete overhaul. Finally, major mining pools even jointly called for a hard fork to resist to the end.

However, Ethereum now has ultra-resistant DeFi, the Beacon chain is up and running, and PoS is ready to be implemented...

Miners want to fork? ¯\_ツ_/¯

Deflation Meme

Another interesting battleground is among Ethereum's Twitter opinion leaders.

In the past, when the Bitcoin and Ethereum communities clashed, the lack of a supply limit for Ether was often mocked by Bitcoin advocates, implying that Ethereum's rampant issuance of coins due to smart contracts and DeFi could not compete with the limited supply of Bitcoin, the digital gold. However, the situation has changed. EIP1559 seems to alter the issuance trend of Ether. If every transaction burns coins, wouldn't this make Ether even more valuable, turning it into a deflationary and ultra-scarce resource? If Bitcoin is sound money, then the post-upgrade Ether could be considered ultra sound money, right?

Thus, Ethereum influencers, like the founders of Bankless and EthHub, began promoting this deflationary meme, suggesting that Ether will take off with DeFi, 2.0, and deflation, eventually surpassing Bitcoin in market value, marked with bat and sound emojis 🦇🔊. However, EIP1559 does not guarantee deflation. While transaction fees are destroyed, block rewards will still mint new coins, potentially more or less. In the long run, the most stable situation would likely oscillate between inflation and deflation.

Some clear-minded individuals choose not to dance with the deflation meme. For example, the founder of MyCrypto stepped in to combat the hype, urging influencers to tone down their claims since Ethereum was already in good shape and didn't need misleading narratives. On the other hand, the developer community remained relatively silent, possibly because price competition with Bitcoin was not their focus. With EIP enthusiasts helping to suppress miners' momentum, they are more concerned with testing chain operations and preparing for the main chain upgrade.

4. Actual Impact of EIP1559

Will Ether Rise?

Unknown.

Death of 0 Gas Transactions

Under the research of Flashbot and ArcherDAO, several schemes emerged that allowed for feeless transactions using MEV searchers. The concept involved setting the gas field to 0 during the transaction but sending funds directly to the miner address (block.coinbase) during contract execution, enabling wallets without Ether but with ERC20 tokens to send transactions.

However, this approach will become impractical after the upgrade, as 0 gas transactions will violate the system's mandatory fee-burning restriction. This may temporarily revert to more original meta transaction relayer methods until future account abstraction solutions are developed.

Automated Fee Setting

This is likely the real intention of EIP1559. After the upgrade, users sending transactions won't need to observe the blockchain mempool status much. They just need to reference the previous block's fee and add a little more, with a high probability of it being executed in the next few blocks. However, for those who want to set fees lower than the current market price and are willing to wait to save money, the time it takes for transactions to be processed will still depend on luck.