Ethereum is a public blockchain that runs programs as well as payments. Developers publish code called smart contracts to it, and anyone can use that code by paying a fee in ether, the network's coin. That design made Ethereum the home of thousands of tokens, including dollar stablecoins, and of trading and lending apps that run without a company in the middle. Working from ethereum.org and live chain data, this guide explains how the system runs, how much ether exists and why that number is rising again, how staking, gas and layer 2 networks fit together, and what can go wrong. For the wider asset class, start at the crypto hub.
What is a smart contract on Ethereum?
A smart contract is a program stored on Ethereum that runs exactly as written whenever a transaction calls it, and every node re-runs it to confirm the result.
Each full node runs the same Ethereum Virtual Machine, or EVM, a standardized computer that executes contract code step by step. Developers usually write contracts in a language such as Solidity and compile them into EVM bytecode. Once deployed, a contract has its own address, can hold ether and tokens, and keeps records such as who owns which balance. Its code cannot be edited afterward unless the authors built in an upgrade switch, and whoever controls that switch becomes someone users have to trust.
Every operation has a price in gas, so a program that loops forever simply runs out of fee money instead of freezing the network. Contracts also cannot read the outside world on their own: a lending app that needs the dollar price of ether relies on a blockchain oracle to post that number on chain.
The applications built this way include:
- Token contracts that follow shared standards such as ERC-20, which is how dollar stablecoins exist on Ethereum (how stablecoins work covers their designs)
- Exchanges and lending pools that run as code, the core of decentralized finance
- Collectibles and ownership records, from NFTs to tokenized funds
Two recent upgrades changed what ordinary accounts can do. Pectra, in May 2025, let a regular key-controlled address borrow smart-contract features such as bundling several actions into one transaction, and Fusaka, in December 2025, added cheap verification for the passkey signatures that phones and laptops already create. Many other blockchains and most layer 2 networks copy the EVM, so the same contract code often runs elsewhere too. The smart contract glossary entry has a shorter definition.
How much ether exists?
Etherscan counted about 122.18 million ETH on November 2, 2026, and Ethereum has no maximum supply.
The total is everything ever issued minus everything burned. Etherscan's breakdown shows where the coins came from:
| Component | ETH |
|---|---|
| Created at launch in 2015: 60 million sold to the public beforehand, about 12 million to the foundation and early contributors | 72,009,990.50 |
| Block rewards paid to miners, 2015 to 2022 | 47,223,894.59 |
| Rewards for uncle blocks, valid blocks that lost a race to be included | 3,139,986.13 |
| Staking rewards since the Beacon Chain started on December 1, 2020 | 4,452,304.20 |
| Base fees burned since August 2021 | −4,642,978.43 |
| Total on November 2, 2026 | 122,183,196.98 |
Two rule changes shaped the recent numbers. The London upgrade on August 5, 2021 began burning the base fee of every transaction, the part of the fee set by the protocol rather than the sender. Then the Merge on September 15, 2022 replaced mining with staking. According to ethereum.org, miners had been receiving about 13,000 ETH a day, and the switch cut new issuance by roughly 88%. Issuance now follows a formula tied to the square root of all staked ETH: more stake raises the total paid out but lowers the rate each validator earns. Etherscan recorded 2,991.6 ETH of new issuance on November 1, 2026.
For about 18 months the burn won. Supply slipped from 120,520,790 ETH on Merge day to a low of 120,065,823 ETH on April 4, 2024, according to Etherscan's daily chart. The yearly totals show how quickly that reversed:
| Year | ETH burned | ETH issued to validators | Net change |
|---|---|---|---|
| 2023 | 1,093,504 | 749,510 | −343,994 |
| 2024 | 634,000 | 926,395 | +292,395 |
| 2025 | 91,175 | 959,325 | +868,151 |
| 2026, through November 1 | 18,942 | 864,436 | +845,494 |
The heaviest single day of burning was May 1, 2022, when 71,718 ETH was destroyed as fees. The heaviest day of 2026 through November 1 burned 1,618 ETH, on October 29.
Why is the ETH supply growing again?
The burn shrank because activity moved to cheaper layers and Ethereum added block space faster than demand grew, so base fees fell close to their floor.
The Dencun upgrade of March 13, 2024 gave rollups a separate, cheap place to post their data, called blobs, and later forks kept raising blob capacity. The block gas limit doubled from 30 million to 60 million during 2025. With more room than users needed, the base fee, which rises when blocks run more than half full and falls when they run less, sank to about 0.044 gwei at the start of November 2026. A gwei is one billionth of an ETH.
Worked example: One day of ETH supply, November 1, 2026
Figures from Etherscan's daily supply chart.
- New ETH paid to validators: 2,991.6 ETH
- ETH burned in base fees: 9.0 ETH
- Net change for the day: 2,991.6 − 9.0 = +2,982.6 ETH
- If every day matched it: 2,982.6 × 365 = 1,088,649 ETH a year, or 1,088,649 ÷ 122,183,197 ≈ 0.89% of the supply
- What the 365 days to November 1, 2026 actually added: +1,001,716 ETH, or 0.83%, because issuance was lower a year earlier (2,660.1 ETH on November 1, 2025) and a few busy days burned more
Worked example: The base fee that would hold supply flat
- Daily issuance to offset: about 2,990 ETH
- Blocks per day: 86,400 seconds ÷ 12 = 7,200, so each block would need to burn 2,990 ÷ 7,200 ≈ 0.415 ETH
- Gas in a block at the target: 30 million, half the 60 million limit
- Base fee needed: 0.415 ETH ÷ 30,000,000 gas ≈ 0.0000000138 ETH, or about 13.8 gwei per unit of gas
- Base fee at the start of November 2026: about 0.044 gwei, roughly 300 times lower
Supply growth of about 0.8% a year is an outcome of current usage, not a fixed policy: a surge in demand for block space would lift the burn, and a drop in staking would lower issuance. Bitcoin vs Ethereum sets this against bitcoin's fixed schedule, and what is bitcoin explains the 21 million cap.
How do validators secure Ethereum?
Validators lock at least 32 ETH, take turns proposing a block every 12 seconds, vote on blocks proposed by others and lose part of their stake if caught cheating.
On November 2, 2026, validatorqueue.com counted 815,282 active validators holding about 44.3 million ETH, roughly 36% of the supply, with a reward rate near 2.45% a year before costs. Since Pectra, a single validator can carry up to 2,048 ETH of effective balance, so one large operator can run a handful of big validators instead of thousands of small ones; validator counts say little about how many independent parties control the stake. About 1.12 million ETH was waiting in the entry queue, a wait of around 20 days, while the exit queue held only 709 ETH.
A block becomes final after two epochs of 32 slots each, about 12.8 minutes, once validators holding at least two-thirds of the stake have voted for it. Undoing a finalized block would cost the attackers a large share of their deposits through penalties, which is the economic core of the design. How blockchain works explains finality in detail and the proof of stake glossary entry gives a short definition. For anyone weighing whether to stake, from running a validator to using a pooled service, what is staking covers the choices and risks, and staking taxes covers how rewards are taxed.
What does it cost to use Ethereum, and where do layer 2s fit?
Fees are counted in gas: a transaction pays the gas it uses times the base fee, which is burned, plus an optional tip for the validator proposing the block.
A plain ETH transfer uses 21,000 gas. At a 0.044 gwei base fee, that is 21,000 × 0.044 = 924 gwei, or about 0.0000009 ETH, under a fifth of a cent at an illustrative $2,000 per ETH. A token swap or a lending deposit uses more gas because the contract does more work, and fees can climb quickly when demand spikes, since the base fee rises automatically whenever blocks run more than half full. The gas fees guide walks through the formula and how wallets set tips.
Layer 2 networks such as Arbitrum, Base and Optimism process transactions away from the main chain, then post compressed batches back to Ethereum in blobs. Blobs have their own fee market, and with capacity raised three times since blobs arrived in 2024, blob fees have stayed very low. That arrangement keeps layer 2 fees down, but it adds new trust points: the operator that orders transactions, the contracts that can be upgraded and the bridge that holds deposited funds. Withdrawing from an OP Stack rollup such as Base back to Ethereum through the standard bridge takes about 7 days. Layer 2 networks explained and crypto bridges explained cover those trade-offs.
Ethereum's upgrades from London to Fusaka
Ethereum changes through coordinated hard forks that every node operator has to install. These six shaped today's network:
| Upgrade | Mainnet date | What changed |
|---|---|---|
| London | August 5, 2021 | EIP-1559 fee market: a protocol-set base fee that is burned, plus tips for block producers |
| The Merge | September 15, 2022 | Proof of stake replaced mining at block 15,537,394 |
| Shapella | April 12, 2023 | Staked ETH could be withdrawn for the first time |
| Dencun | March 13, 2024 | Blobs for rollup data, with a target of 3 and a maximum of 6 per block |
| Pectra | May 7, 2025 | Smart-account features for ordinary addresses (EIP-7702), validators of up to 2,048 ETH, blob target raised to 6 |
| Fusaka | December 3, 2025 | PeerDAS, which lets nodes check blob data by sampling pieces of it; passkey signature support; a cap of 16,777,216 gas per transaction |
Fusaka also set up blob-parameter-only forks, small upgrades that change nothing but blob limits. The first, on December 9, 2025, raised the target to 10 blobs per block, and the second, on January 7, 2026, to 14, with a maximum of 21. Separately, validators raised the block gas limit by signaling in the blocks they proposed: from 30 million to 36 million in February 2025, to 45 million in July 2025 and to 60 million on November 25, 2025, eight days before Fusaka itself. The next named upgrade, Glamsterdam, was still listed on ethereum.org without a date on November 2, 2026.
Ether's price record
Ether trades around the clock, and the latest price for one ETH is $2,287.65 (Nov 2, 2026).
Its history is one of steep climbs and deep falls. Ether lost about 94% of its value during 2018, and from its August 2025 high to its June 2026 low the daily close fell about two-thirds, according to the FRED-based figures in bitcoin vs ethereum, which also compares its volatility with bitcoin's year by year.
| Asset | Last close | 1 year | 3 years (per year) | 5 years (per year) | 10 years (per year) | Worst drop from a high | Volatility (1 year) |
|---|---|---|---|---|---|---|---|
| Ethereum | $2,309.83 | -40.9% | +8.7% | -12.8% | n/a | -93.3% (Dec 14, 2018) | 60.8% |
Treat the table as a record of what holding ether has demanded, not as a guide to what comes next. Each return column is a compound yearly rate between two dates, so a ten-year figure that starts when ether traded far lower can look nothing like the three-year one. The worst-drawdown column shows the deepest fall from a previous high, the number to weigh against any return figure, and the volatility column measures how wide the daily swings were over the past year. When the table cannot load, the ethereum price page still shows the current chart.
Ways to own ether include buying the coin on a trading platform and keeping it there or in a wallet you control, or buying a spot ether ETF, a type of fund that has traded in the US since July 2024. How to buy ethereum and ether ETFs explained compare the routes, and custodial vs non-custodial wallets explains who holds the keys in each case.
What are the risks of owning ether?
Ether is volatile and can lose most or all of its value, and holding it brings technical, policy and custody risks that a bank deposit does not carry.
- Price risk. The falls described above happened while the network kept working normally. No earnings or interest payments put a floor under the price.
- Smart-contract risk. Code that holds money can be exploited. In 2016 an attacker drained more than 3.6 million ETH from a contract called The DAO, and the community answered with a hard fork at block 1,920,000 that moved the funds; those who rejected the fork kept the original chain, now Ethereum Classic. Apps built today carry the same kind of risk, with no rescue on offer.
- Changing supply. Issuance currently exceeds the burn by about 0.8% a year, and future upgrades can change the economics again.
- Staking risk. Penalties, queues that can stretch for weeks and pooled-staking tokens that can trade below the ETH they represent; see what is staking.
- Layer 2 and bridge risk. Rollups depend on operators and upgradeable contracts, and bridges concentrate deposits that attract hackers.
- Mistakes and scams. Sending ETH on the wrong network or to a mistyped address is usually permanent (sent crypto to the wrong network), and fake token and airdrop schemes are common (crypto scams).
- Rules can change. The SEC's March 2026 interpretation treats ether as a digital commodity, but no market-structure law had passed Congress as of November 2, 2026 (US crypto regulation).
One way to size a holding is to ask what a fall of 80% would do to your other plans; how much crypto to hold works through that question. Chain and staking figures on this page were checked on November 2, 2026.