After a strong bear market, Ethereum has been back in vogue. Due to a confluence of factors — including a rally in the ETH price and adoption of so-called "decentralized finance" (DeFi) — the blockchain has seen a strong increase in active users and daily transactions.
This spike in usage has corresponded with an uptick in transaction fees, now reaching levels not seen in years. While transacting ETH from wallet to wallet only costs $0.10 (this figure is current as of July 3) according to ETH Gas Station, interacting with decentralized applications has become a costly affair — something that analysts say is a threat to Ethereum's viability.
These high transaction costs come in spite of a potentially risky short-term solution that was implemented by miners. Long-term and more stable solutions are needed to ensure Ethereum's viability.
One entrepreneur in the space, in fact, went as far as to say that as long as fees remain at high levels, Ethereum can be overtaken or "dethroned" by other smart contract blockchains.
Growth in DeFi, Stablecoins Met With A Surge in Ethereum Transaction Costs
In 2017 and 2018, Ethereum seemingly had one leading purpose: it was a platform for entrepreneurs to launch initial coin offerings (ICOs). After a collapse in the popularity of ICOs, decentralized finance has filled the role of Ethereum's leading use case.
Simply put, DeFi is a segment of the cryptocurrency space where developers attempt to migrate traditional financial services onto a decentralized framework.
Due to its pre-existing user base, a relatively easy coding language, and network effects, decentralized finance has found a home on Ethereum. DeFi's poster-children in Compound, MakerDAO, Synthetix, and the Kyber Network are all based on the blockchain. And the network of choice for stablecoins such as USD Coin (USDC) and Tether (USDT) is Ethereum.
With billions of dollars worth of stablecoins being issued on Ethereum as DeFi has seen a strong uptick in adoption, Ethereum has naturally grown.
Blockchain analytics firm Santiment reported on June 23 that there are now over 100,000 ETH addresses created every day. And Etherscan reports that every day since June 22, there have been approximately one million confirmed Ethereum transactions. Such a consistently high number of confirmed Ethereum transactions hasn't been seen since the peak of the 2017 and 2018 bubble.
Growth is welcome, sure: an increase in Ethereum's active user count and viable use cases is healthy. But, expectedly, it has resulted in transaction costs skyrocketing.
Like Bitcoin, the cost of sending data — meaning sending ETH, interacting with a smart contract, or otherwise — on Ethereum is set by the supply-demand dynamics of a free market: the more demand for something there is, the more that something will cost.
Where Bitcoin has satoshis per byte, Ethereum has gwei per gas. And as a satoshi is one-hundred-millionth of a BTC, a gwei is one-billionth of an ETH.
Due to the recent spike in transaction demand, the amount of gwei Ethereum users are willing to spend for gas, for transactions has reached ludicrous levels.
Blockchain analytics and data service provider TradeBlock reported that for a few days in mid-June, gas costs were as high as 125 gwei. Per the company's data, this is the highest gas costs have been in over three years.
At 125 gwei and an ETH price of $225, most interactions with Ethereum become unsustainable, warranting solutions:
Sending ETH to another address, a 21,000 gas operation, costs $0.59.
Sending an ERC-20 token like USDT, a 52,146 gas operation, costs $1.46.
Depositing cryptocurrency into the most popular DeFi application Compound, a 200,000-400,000 gas operation, can cost upwards of $5.
Trading on decentralized liquidity protocol Kyber Network, reportedly a 500,000+ gas operation, can cost upwards of $10.
Fees Are Becoming Unbearable for Users
Even though the 125 gwei spike was a temporary blip, gas costs remained high days later.
The image below shows the daily aggregate of transaction fees paid on Ethereum denominated in ETH.
Stretching back to the blockchain's launch in 2015, the data shows that miners have collected over 2,500 ETH a day in transaction fees for the past month or so. 2,500 ETH in transaction fees is far from an all-time high, but this is the longest period of high transaction fees the blockchain's users have ever seen.
Furthermore, as of July 1, the average gas cost is 47.36 gwei, according to Etherscan data.
The high cost of sending transactions has also been confirmed anecdotally.
The head of business development at Kraken’s futures division, Kevin Beardsley, said in June:
"I have spent $14 on ETH gas fees to transfer/lock my $15 into @CurveFinance and I’m earning a princely $0.079 in weekly $SNX rewards. I’ll break even in just 177 short weeks! (not including gas to close contracts.”
Qiao Wang, a former head of product at Messari, has echoed this frustration. Referencing how he spent $10 to interact with a DeFi smart contract, the analyst wrote:
“So long as ETH 2.0 is not fully rolled out, there’s an obvious opportunity for a highly scalable blockchain to dethrone Ethereum. Paying $10 transaction fee and waiting 15 seconds for settlement is just bad UX.”
Because DeFi transactions can cost a handful of dollars each, relatively small investors in and users of Ethereum and tokens based on the network are feeling the brunt of high transaction fees. If it costs $5 for one to earn a return on their $100 worth of Ethereum, it makes little sense to go ahead with this transaction. The same could be said about even larger transactions.
The Short-Term Solution Didn't Work
The persistence of these high fees come in spite of the introduction of a vote amongst Ethereum miners to increase the gas limit of each block from 10,000,000 to 12,500,000. The proposal was successfully voted on and passed on June 19, giving the network more transaction capacity. (As a pertinent aside, the gas limit remains capped at 12 million, for the time being, presumably to ensure the network stays stable.)
With gas costs still high and users still raving about the inefficiency of Ethereum transactions, this short-term solution did not work.
Worse yet, some commentators say that increasing the gas limit of Ethereum blocks is a dangerous affair
Péter Szilágyi, a team lead at the Ethereum Foundation, bluntly said in the aftermath of the vote that "Ethereum miners don't give a fuck about the long term health of the network nor about DoS attacks."
Anthony Sassano, the author of The Daily Gwei, broke down those concerns further in a June 22 edition of the newsletter.
Because the gas limit is basically a proxy for how much data is being sent to and processed by miners, an increase to this parameter puts pressure on miners. Miners that may be running weaker hardware or those with a slow connection may process blocks slower, be more susceptible to a denial of service attack, and will have to spend more on storing Ethereum blockchain data.
As it stands, these fears are just theoretical: Ethereum continues to process blocks without much of a hitch. The point that Sassano, Szilágyi, and others were trying to make is that this change increases the risk to the health of miners.
What Solutions Are There?
With miners implementing a potentially dangerous solution to Ethereum fees that did not fully work, what long-term and more safe solutions are there?
Firstly, there is Ethereum Improvement Proposal 1559 (EIP-1559). Authored by blockchain founder Vitalik Buterin and other developers in the space, the proposal outlines a system that will remove the "Inefficient" and unstable auction model that currently decides transaction fees. EIP-1559 would also abolish the rights miners have to change the gas limit; changes to the gas limit will be a capacity managed by hard forks instead.
Secondly, there are second-layer scaling solutions. Like Bitcoin's Lightning Network, Ethereum has its own developers trying to mitigate the brunt of transactions to a secondary secure network. Such systems allow for faster transactions, cheaper transaction fees, and potentially more security and privacy. Some of Ethereum's second-layer solutions are Skale Network, OmiseGo, Starkware, and MATIC Network. Few of these solutions have seen the adoption of the mainstream, though, yet many see them as critical for Ethereum scaling.
And lastly, there is Ethereum 2.0 — a full-fledged upgrade to the blockchain that will fundamentally change how it operates. The upgrade has been in the works for years, and it is expected to dramatically increase Ethereum's transaction capacity, thus decreasing fees and increasing usability.
The first phase of this upgrade is expected to go live in 2020, according to developers working on Ethereum 2.0 code and integrations. Founder of the blockchain, Vitalik Buterin, has confirmed this himself. Unfortunately for those worried about high fees, the first phase is only expected to activate a small portion of Ethereum 2.0's technology.
The full rollout of the upgrade, which will largely mitigate the ongoing transaction fee issues, will take over a year. And some entities think it will take even longer than that. BitMEX Research once commented:
“Ethereum 2.0 is exceptionally complicated. With so many committees, shards and voting types it seems reasonably likely that something will go wrong and that there will be significant further delays.”