Who miners and validators are, and how they get paid

The parties that produce blocks are businesses. They have electricity contracts, hardware leases, colocation bills or large amounts of capital locked in a protocol, and they earn from two or three identifiable revenue lines. Understanding those lines explains most of what block producers do, including the behaviors that look strange from outside.

The job description

A block producer collects pending transactions, chooses which to include and in what order, assembles them into a valid block, and broadcasts it to the network fast enough that other nodes build on it rather than on a competitor's. Under proof of work the right to publish is won by a continuous hashing race, so the producer is called a miner. Under proof of stake the right arrives on a schedule determined by bonded stake, so the producer is called a validator. In both cases the discretionary part of the job is the same: selecting and ordering the contents of the mempool.

Neither role is a position of authority over the ledger. A producer who includes an invalid transaction produces a block that every full node discards, along with the reward. What producers control is inclusion and sequence, which is exactly the domain in which their commercial incentives operate.

Why almost nobody does this alone

Block rewards arrive at random. A small miner whose share of network hash power is one part in a hundred thousand will find a block roughly that often, which means years between payouts and enormous income variance. Mining pools solve this by combining hash power, taking a fee, and paying members a smooth share proportional to work contributed. The consequence is that the visible producer of a block is often a pool, and the underlying machines belong to many parties who can switch pools.

Proof of stake has an analogous structure. Delegation and staking services let holders assign stake weight to professional operators for a commission, and liquid staking providers aggregate deposits at scale. The result in both systems is that economic ownership is dispersed while operational control is concentrated, which is why the Nakamoto coefficient is reported separately from validator counts, and why validator concentration is tracked as a distinct risk.

The cost structures behind the two roles are different in a way that shapes behavior. A miner's hardware is a sunk cost and electricity is the marginal one, so a machine keeps running as long as the expected reward exceeds the power it burns, and it is switched off the moment that stops being true. That is why proof-of-work capacity migrates toward the lowest-cost power and why capacity leaves the network in waves rather than smoothly. A validator's main cost is the capital locked in the bond, which is not consumed but is unavailable for anything else while it is staked, plus a comparatively small operational bill for a server that must stay online. The first business is exposed to energy prices; the second is exposed to the terms of the lock-up and to slashing.

The three pay lines

The first is the block subsidy: newly created units, paid by the protocol on a published emission schedule. Nobody hands over this money. It is issuance, so it is funded by dilution of existing holders, and it appears in data as daily issuance value.

The second is transaction fees, which users actually pay. On chains using EIP-1559 style pricing, the fee splits into a base fee that adjusts algorithmically with congestion and is destroyed rather than paid to anyone, and a priority fee that goes to the producer. That split matters: on such a chain, a large share of what users pay never reaches the parties securing the network at all, and instead reduces supply. Daily fees and daily burn report the two halves.

The third is MEV, the value obtainable purely from choosing the order of transactions. It includes benign forms such as capturing an arbitrage that a pending trade creates, and predatory forms such as front-running and the sandwich attack, where a trader's own transaction is bracketed to profit from the price it moves. On several networks this has produced a specialized supply chain: searchers find profitable orderings, block builders assemble full blocks, and proposers auction the right to publish to the highest-bidding builder. MEV is real income for producers and a real cost to some users, and it is only partially observable in public data.

What this is called, and what the words do not mean

Aggregate payments from users to producers are reported here as supply-side revenue, and the term needs care because it borrows an equity word. Corporate revenue accrues to a company, which has costs, residual profit and shareholders with a claim on it. Supply-side revenue accrues to a competitive, open set of operators who are not owned by holders of the asset and who distribute nothing to them. A rising figure means users paid more for block space; it does not mean an entity earned more on behalf of anyone.

Where fees are burned or diverted to a protocol treasury, the flow is closer to accruing to holders, and it is reported separately as holder revenue rather than being mixed into a single number. Even then the analogy is loose: a burn reduces supply for every holder proportionally and creates no cash, no distribution, and no enforceable claim. Ratios built from these figures, such as market cap to fees, resemble equity multiples in construction only. They have no denominator that anyone receives.

Reading a producer economy

Four figures describe it compactly: issuance for the subsidized part of the pay, fees for the user-paid part, supply-side revenue over thirty days to smooth out daily noise, and hash rate or staked supply for the resources those payments are supporting. The relationship between the pay and the resource is the subject of the next lesson.

For per-network detail, see the fee pages and the staking pages, and consult methodology for how MEV and burned fees are classified in these figures.