Fee Markets & EIP-1559

Block Space Is Scarce — So It's Auctioned

Every block can only hold so many transactions, so block space is a scarce resource, and scarce resources get priced. That price is the transaction fee (gas fee) you pay to get included. When the network is busy, more people compete for the limited space, and fees rise; when it's quiet, fees fall. In effect, a blockchain runs a continuous auction for block space, and your fee is your bid to be included. Understanding fees means understanding this market: you're not paying a fixed 'postage', you're bidding against everyone else who wants their transaction in the next block, and the busier it is, the higher the going rate for the space you're competing for.

The Old Way and Its Problems

Before EIP-1559, Ethereum used a simple 'first-price auction': everyone submitted the fee they were willing to pay, and block producers picked the highest bidders. This worked but had annoying problems. Fees were hard to estimate — you had to guess what bid would get you in, and people routinely overpaid (bidding far more than necessary out of caution) or underpaid (and got stuck waiting). The experience was volatile and confusing, especially during congestion, with wildly swinging fees and frequent overpayment. EIP-1559, activated in 2021, was a major redesign of this fee market aimed at making fees more predictable and the mechanism more efficient, while introducing a notable economic change: burning part of the fee.

• Block space is scarce, so transaction fees are essentially bids in an auction
• Busier network → more competition → higher fees; quiet → lower fees
• The old first-price auction made fees hard to estimate and prone to overpaying
• EIP-1559 (2021) redesigned Ethereum's fee market for predictability

How EIP-1559 Works: Base Fee + Priority Fee

EIP-1559 splits the fee into two parts. The base fee is a network-set price per unit of gas that adjusts automatically block by block: if the previous block was more than half full, the base fee rises; if less than half full, it falls. This algorithmic adjustment targets blocks being about half full on average, so the base fee moves toward a market-clearing level and becomes far more predictable than guessing in a blind auction. On top of the base fee, users add a priority fee (a 'tip') to incentivise block producers to include their transaction preferentially — useful when you want to jump ahead during congestion. So your total fee is (base fee + priority fee) × gas used. The base fee handles the 'what does space cost right now' question transparently; the tip handles 'how urgently do I want in'.

The Burn: Where the Base Fee Goes

EIP-1559's most consequential twist is what happens to the base fee: it is burned — permanently destroyed — rather than paid to the block producer. Only the priority-fee tip goes to the producer. Burning the base fee was designed partly to prevent block producers from gaming the fee mechanism (they can't profit from inflating the base fee if they don't receive it), and it has a large monetary side effect: it removes ETH from circulation in proportion to network usage. When the network is heavily used, more ETH is burned, which can offset or exceed new issuance, making the asset net-deflationary during busy periods. This burn is why EIP-1559 is discussed as much for its monetary policy implications as for its fee-market mechanics — it tied the cryptocurrency's supply dynamics directly to how much the network is used.

What EIP-1559 Did and Didn't Fix

It's important to be precise about EIP-1559's effects. It did make fees more predictable and the fee-setting experience smoother (wallets can reliably estimate the base fee), reduced overpayment from blind bidding, and introduced the burn. What it did not do is make fees cheap: EIP-1559 is a better pricing mechanism, not a scaling solution. When demand for block space is high, the base fee still rises — sometimes sharply — because the fundamental scarcity is unchanged; the mechanism prices congestion more smoothly but cannot conjure more space. Lowering fees fundamentally requires more capacity (layer 2s, blobs, larger effective throughput), not a better auction. Conflating EIP-1559 with 'cheaper Ethereum' is a common misunderstanding: it improved how the price is discovered and where the fee goes, not the underlying supply of block space that sets the price level.

• Base fee: network-set, auto-adjusts each block toward ~half-full blocks (predictable)
• Priority fee (tip): extra incentive to be included preferentially during congestion
• The base fee is burned (destroyed), not paid to producers — anti-gaming + supply effect
• EIP-1559 improved predictability and monetary policy, not the cost/scarcity of space

Multidimensional Fees and the Blob Market

Fee-market design has grown more sophisticated because not all block resources are the same. A transaction consumes several distinct resources — computation, storage access, and data — and pricing them all with one gas number is a crude approximation that can misprice usage and create odd incentives. Multidimensional fee ideas price different resources separately so each is charged according to its own scarcity. The clearest realisation is the blob fee market introduced with proto-danksharding (EIP-4844): rollups post their data as 'blobs', which have their own separate fee market and base fee (with its own EIP-1559-style adjustment and burn) distinct from regular gas. This means the price of rollup data availability floats on its own supply and demand, largely independent of ordinary execution fees (since the Fusaka upgrade of December 2025, EIP-7918 adds a reserve price to blobs that is proportional to the execution base fee). It's a direct extension of the EIP-1559 mechanism to a new, separately-scarce resource, and it reflects a broader move toward pricing each blockchain resource on its own curve rather than bundling everything into one fee.

Fee Markets, MEV, and Cross-Chain Design

Fee markets don't exist in isolation — they interact with MEV and block-building. The priority-fee tip is only one way producers earn from ordering; MEV (reordering, sandwiching, arbitrage) is another, and sophisticated block-building optimises across both. This is why fee-market design connects to proposer-builder separation, private order flow, and the broader question of how value from block space is captured and distributed. Different chains also make different fee-market choices — some use variants of EIP-1559, some use local fee markets (pricing congestion per-application or per-account to stop one hot contract spiking fees for everyone), and some use entirely different models — reflecting different priorities around predictability, throughput, and fairness. Understanding a chain's fee market therefore means understanding not just its base-fee mechanics but how tips, MEV, and any local-fee-market design shape who pays what and who captures the value of scarce space.

A Systems View of Fees

For a sophisticated observer, fee markets are best seen as the economic layer that allocates a blockchain's scarce resources, and EIP-1559 as a landmark in making that allocation more efficient, predictable, and monetarily meaningful (via the burn) — without changing the underlying scarcity. The mature understanding holds several things at once: fees are auction prices for scarce space; EIP-1559 improved price discovery and tied supply to usage but is not a scaling fix; multidimensional and blob fee markets extend the idea to price distinct resources separately; and fee dynamics are entangled with MEV and block-building. Reading a chain's fees this way — as a designed market with mechanisms, trade-offs, and interactions, rather than a flat 'cost of using it' — clarifies why fees behave as they do, why 'cheaper fees' ultimately requires more capacity, and how design choices around base fees, burns, tips, and resource pricing shape both user experience and the asset's economics. It is one of the clearest places where cryptoeconomics and everyday user experience meet.