The lack of scalability on Ethereum is a common talking point in crypto spaces. When gas prices are spiking, people are quick to grab their pitchforks and claim the system is broken. But the reasoning and economic justification for expensive gas fees is often overlooked or brushed over. Such criticism misses a couple key insights:

• Transaction fees onchain do not scale with size.

• Gas prices fluctuate dynamically because of a changing demand for a limited supply of blockspace.

In traditional financial markets, such as the stock market or even CEXs, fees are percentage based, and as a result, directly scale with size. Onchain, this is not the case. Users pay the same transaction fees to validators regardless of size. In decentralized consensus mechanisms, compute becomes expensive. As a result, onchain fees are based on entirely different parameters:

• how much gas the transaction consumes.

• the current gas price.

Gas consumption per transaction can be understood, at a high level, as how complex the transaction is. Each action performed onchain costs gas. Some transactions are very simple, such as sending ETH from one wallet to another. Other transactions, such as flash loans, can be very complex, and as a result, cost more gas.

This mechanic is a direct byproduct of the decentralization of consensus mechanisms. In traditional finance, everything is handled by a centralized consensus mechanism, also known as a server. As a result, compute is very cheap. But for decentralized consensus mechanisms, compute becomes very costly. As consensus needs to be achieved, state updates are handled across a multitude of validators, which comparatively takes a lot more time and effort.

Originally coined by Vitalik Buterin, The Scalability Trilemma states that all consensus mechanisms would like to be scalable, secure, and decentralized. But, only two out of three are possible, since each option comes with its own unique set of tradeoffs. Payment processors such as Visa are fast, scalable, but centralized. Consensus mechanisms such as Ethereum are decentralized, secure, but not scalable... as indicated by how expensive gas fees can get during times of peak network activity.

It is, in fact, due to the tradeoffs made in favor of decentralization and security, unavoidable, that gas prices on Ethereum can get so expensive. Onchain gas prices only become expensive when users are willing to pay a premium for blockspace! If a user pays an expensive price to ensure their transaction lands on the next block, they are doing so because they believe the cost is worth the opportunity in return.

During the most recent on chain meme frenzy where $PEPE rushed to a billion dollar valuation in a matter of weeks, gas prices spiked. This is nothing but expected, and economically, makes sense! Since users were rushing to buy/sell meme coins on dex markets, they were willing to pay a pretty penny to ensure their transaction is included in the next available block. After the meme coin bubble died down, the cost for transactions on Ethereum quickly followed suit. This is only natural as there is no longer a demand for blockspace.

Onchain gas economies are an unavoidable tradeoff of decentralized consensus mechanisms. Visa processes 20,000 transactions per second with ease and at low cost because compute is handled on one big server that doesn't need to worry about consensus. Blockchains such as Ethereum care about consensus, and as a result, are forced to sacrifice and limit throughput. Blockchains that claim to have, "solved the blockchain trilemma", should always be treated with skepticism for this reason.

The auction market dynamics of gas prices on Ethereum are a necessary evil, and a direct cause-effect result of the need to reach consensus. When throughput is scarce, it becomes a commodity. In the same way that social events cannot have inclusivity without exclusivity, blockchains cannot limit throughput without creating a demand for blockspace.

P.S. It is worth mentioning that there are proposed workaround solutions out there such as sharding/rollups that do not claim to be pulling any rabbits out of hats such as other blockchains that claim to solve the scalability trilemma that will not be mentioned, but there is a lot to discuss on that front, better suited for another article :3