A user in early 2024 attempts to swap tokens on Ethereum mainnet using their self-custody wallet. The quoted gas fee is $47. They switch to Arbitrum, initiate an identical transaction, and see a fee of $0.12. The same operation, the same wallet interface, the same action—yet the cost difference is nearly 400-fold. This is not an error in the wallet’s display, nor is it pricing variance across exchanges. It is the direct result of how different blockchains structure transaction costs, validator economics, and network capacity.

Rabby Wallet, designed specifically for Ethereum and EVM-compatible networks, makes this reality visible in a way that generic multichain wallets often obscure. When a user connects to Ethereum mainnet versus Layer 2 solutions like Arbitrum, Optimism, or Base, the wallet displays different fee structures, confirmation times, and network conditions. Understanding why those differences exist is essential for anyone managing assets across multiple blockchains. The variation is not arbitrary; it reflects fundamental technical choices about how each network validates transactions, allocates block space, and compensates validators.

How Ethereum mainnet gas pricing works

Ethereum mainnet uses a continuous auction model for block space. Every transaction specifies a maximum fee per unit of gas (the “max fee per gas”) and a priority fee that goes directly to validators as a tip. The network processes roughly 12-13 million gas per block, with blocks produced every 12 seconds. When demand exceeds this capacity—during market volatility, NFT mints, or popular token launches—users compete by offering higher fees. The protocol automatically calculates a “base fee” that adjusts every block based on whether the previous block was fuller or emptier than the target, creating a feedback loop that makes Ethereum’s fee market responsive but unpredictable.

This design has a crucial consequence: gas costs scale with network congestion, not with transaction complexity alone. A simple token transfer and a complex smart contract interaction both consume discrete units of gas, but neither is “expensive” in an absolute sense. Expensiveness emerges from competition for limited block space. When Ethereum validators fill blocks near capacity, the base fee rises, and transaction costs can climb rapidly. During periods of lower activity, fees can fall below $1 per transaction. Rabby Wallet’s transaction simulation feature accounts for this by showing the expected balance change and fee at the current network state; if network conditions change between quote and confirmation, the fee can be different.

The base fee is also algorithmically destroyed, not given to validators. Only the priority fee (the tip) reaches the block proposer. This separation was introduced in Ethereum Improvement Proposal 1559 to reduce the unpredictability of fee markets and make it easier for users to set fees without guessing the true market price. However, it also means that when network usage is high, a substantial portion of transaction costs effectively disappears from the economic value chain—it is not paid to validators, developers, or service providers. It is simply removed from circulation. For users paying significant Ethereum fees, this is neither comforting nor costly; it is simply the mechanism.

Ethereum mainnet’s security model depends on this fee market. Validators are paid through transaction fees and block rewards (which are being phased down). If transaction fees were artificially suppressed or if the network accommodated unlimited transaction volume, validator revenue would decline, and the economic incentive to run a validator node would weaken. The high fees observed on Ethereum during periods of demand are not a failure; they are a signal that the network is at capacity and that additional throughput would require architectural changes.

Layer 2 networks reduce costs through batching and compression

Arbitrum, Optimism, Base, and other Layer 2 solutions process transactions with fundamentally different economics. These networks bundle thousands of transactions together and post a summary to Ethereum mainnet roughly once per minute (or even less frequently, depending on the specific implementation). Instead of every transaction requiring separate validation by Ethereum’s full validator set, Layer 2s maintain their own sequencers or validators that process transactions and then compress the results into a single state update posted to the main chain.

The cost difference is enormous because Layer 2s only pay Ethereum gas fees for the compressed data they submit, not for every transaction individually. If a Layer 2 sequencer posts a batch containing 1,000 simple transfers, the Ethereum gas cost is split across all 1,000 transactions. Each participant’s share of the mainnet posting cost becomes negligible. Additionally, Layer 2s use data compression techniques to reduce the size of information posted to Ethereum. Optimistic rollups (used by Optimism, Arbitrum, and Base) compress transaction data further by relying on fraud proofs—if a state transition is invalid, any observer can submit a proof to Ethereum showing the error, and the Layer 2’s state is rolled back.

This architecture creates a tiered fee structure. Layer 2s still require gas for on-chain execution—validators need to compute state changes, maintain local storage, and process transactions. But because that execution happens off-chain and is not directly replicated across Ethereum’s thousands of nodes, the computational cost per transaction is far lower. A typical Layer 2 transaction might cost $0.05 to $0.50, compared to $15 to $150 on Ethereum mainnet during busy periods. Rabby Wallet displays these differences explicitly by allowing users to switch networks and see real-time fee estimates for the same action on different chains.

The trade-off is security and latency. Ethereum mainnet settles transactions with finality in roughly 13 minutes (two epochs of the Beacon Chain). Layer 2 transactions are confirmed by the Layer 2’s sequencer almost immediately, but final settlement to Ethereum can take days (in the case of fraud proofs that need to be challenged) or requires trust in a bridge operator. Users who need cryptographic certainty that a transaction cannot be reversed must wait for Layer 2 state to be confirmed on Ethereum, which can take hours or days depending on the configuration.

Why different networks have different validator economics

Polygon, Avalanche, BNB Chain, and other EVM-compatible networks operate independent blockchains with their own validator sets and security models. They do not post transaction summaries to Ethereum; instead, they run parallel consensus mechanisms. This independence allows them to set their own gas prices based on their own network capacity and validator requirements.

Polygon’s validators operate with much lower hardware requirements and economic rewards than Ethereum validators. Ethereum validators must stake 32 ETH (currently worth over $100,000) to participate, and network security is backed by the full value of Ethereum’s token supply and the economic incentives of its ecosystem. Polygon validators can stake less, and the network’s security model is different. Consequently, Polygon can operate with lower transaction fees while still compensating validators adequately. The trade-off is that Polygon’s security guarantees are different from Ethereum’s. Polygon transactions are final after a shorter period and depend on the security of Polygon’s validator set, not Ethereum’s consensus.

BNB Chain follows a similar model but with centralized governance and a smaller validator set. Its fees are typically lower than Polygon’s, but its security model is more centralized. Avalanche uses different consensus entirely, with different throughput, finality, and fee characteristics. The important pattern is that lower fees do not simply mean “better”—they mean the network has chosen a different security model, validator economics, and capacity design. Rabby Wallet’s multichain portfolio viewing makes these trade-offs visible by allowing users to see the same assets on different networks and understand the cost implications of moving between them.

How Rabby displays network selection and fee estimation

When a user opens Rabby Wallet and connects to a decentralized application or initiates a transaction, the wallet’s automatic network selection identifies which EVM network the application expects. If the user’s wallet is currently set to Ethereum mainnet but the dapp requires Arbitrum, Rabby can suggest switching networks. This automation reduces the most dangerous error in multichain wallets: submitting a transaction on the wrong network, which can result in funds being sent to addresses that do not exist on that chain or lost entirely.

The transaction simulation feature shows the expected balance changes before the user signs. This is crucial for understanding fees. When a user approves a swap or transfer, Rabby simulates the transaction against the current network state and shows the resulting asset balances. If the swap will cost $50 in fees on Ethereum but $0.20 on Arbitrum, that difference appears in the simulation. Users can see not just the headline fee amount, but the impact on their final balance. If slippage, price impact, or unexpected execution paths are part of the transaction, simulation can sometimes reveal those as well, though complex smart contract interactions may not always be fully predictable.

Approval visibility is another critical feature for understanding network costs and risk. When a smart contract requests permission to spend a user’s tokens (through the ERC-20 approve function), Rabby displays what the contract is being approved to do and how many tokens it can access. This matters for fees because approvals themselves are transactions that consume gas. On Ethereum mainnet, approving a new contract might cost $10 to $30. On Arbitrum, the same approval might cost $0.05. Understanding approval costs helps users decide whether to approve contracts that they plan to use frequently (where the per-transaction amortized cost is lower) or use permit-style approvals that bundle approval and execution into one transaction, saving a step but not necessarily reducing total gas.

The practical impact of gas fee differences on portfolio strategy

High mainnet fees create strong incentives for users to conduct frequent trading and portfolio management on Layer 2s and alternate chains. If a user makes five token swaps per week, the cumulative fee difference between Ethereum mainnet and Arbitrum could be hundreds of dollars annually. This economic pressure has real consequences: users cluster their activity on cheaper networks, which creates liquidity fragmentation. The most liquid markets remain on Ethereum, but users conducting smaller transactions move to Layer 2s. Arbitrage traders, large liquidity providers, and frequent traders pay the highest fees on Ethereum; casual users and small-balance holders migrate to cheaper networks.

This fragmentation also affects how users bridge assets between networks. Moving assets from Ethereum to Arbitrum requires using a bridge, which typically involves depositing on Ethereum (paying mainnet gas) and withdrawing on Arbitrum (paying a Layer 2 fee). For a small transfer, the combined cost might be $50 or more. Users therefore tend to move larger batches of assets less frequently, which creates operational friction. Rabby Wallet’s unified multichain portfolio viewing makes this trade-off visible: a user can see their assets scattered across Ethereum, Arbitrum, Optimism, and Base, and understand that consolidating everything to one network would require paying bridge and gas fees.

Long-term holders and passive participants benefit most from understanding network economics because they can optimize entry and exit points. If you plan to hold a position for a year, paying a $0.10 approval and swap fee on Arbitrum versus $30 on Ethereum is a massive difference in final returns. Conversely, if you need to exit a position immediately during a market event, the cheapest network might not have sufficient liquidity, forcing you to use Ethereum mainnet despite the cost.

One often-overlooked detail is that Layer 2 fees are not static. As Layer 2 networks grow and post more data to Ethereum, the per-transaction share of mainnet posting costs can increase. If Arbitrum’s transaction volume doubles, the cost per transaction may rise by 20-30% because the per-byte cost of posting to Ethereum is constant and is now shared among more transactions. Conversely, if mainnet Ethereum base fees increase dramatically, Layer 2 fees increase as well because the Layer 2 must pay more to post its compressed data. Rabby Wallet’s gas estimates are real-time because they must account for these dynamic conditions.

Understanding approval costs and smart contract interactions

Smart contract interactions often require two transactions: an approval and the actual action. Approving a decentralized exchange to spend your tokens, approving a lending protocol to take collateral, or approving a bridge to move assets across networks all require separate approval transactions. On Ethereum mainnet during high-congestion periods, each approval might cost $20 to $50. On Arbitrum, the same approval costs $0.10 to $0.30.

This creates a genuine cost-benefit calculation. If you plan to use a contract multiple times, paying for a single approval is efficient because subsequent transactions do not need another approval. However, if you use a contract once and never return, you have paid for an unnecessary transaction. Some protocols use permit-style approvals (EIP-2612) that bundle the approval and the action into a single transaction, eliminating the separate approval step. Rabby’s approval visibility helps users identify which approach is being used and make informed decisions about network selection based on the total cost.

Smart contract execution costs vary less dramatically across networks than approval costs, because execution depends primarily on the contract’s computational complexity and storage access, not on network congestion. A swap on Uniswap consumes roughly the same amount of gas on Arbitrum as on Ethereum mainnet. But because Layer 2s compress transactions before posting to Ethereum, and they do not require all validators to recompute every transaction, the per-user cost is much lower. This is why the same dapp feels dramatically cheaper on a Layer 2 even though the underlying smart contract logic is identical.

Bridging assets and the true multichain cost

Moving assets from Ethereum to Arbitrum or Optimism requires using a bridge, which is itself a smart contract interaction that consumes gas on both sides. A typical bridge operation costs $20 to $60 on Ethereum (depositing into the bridge) and $0.50 to $5 on the Layer 2 (receiving the bridged asset). For a $500 transfer, the total cost is 4% to 12% of the amount moved. For a $50,000 transfer, the cost drops to under 0.15%.

This creates a natural minimum transaction size below which bridging is economically irrational. If you want to move $1,000 from Ethereum to Arbitrum, paying $40 to do so eats 4% of your capital. If you want to move $100,000, the same fee is only 0.04%. Large institutional users and high-frequency traders can afford to bridge frequently because the per-transaction cost is amortized across large balances. Small retail users either accept high percentage costs or consolidate their assets on one network.

Some bridges offer alternative settlement options, such as liquidity pools that allow faster, cheaper exits from Layer 2s back to Ethereum at the cost of accepting slippage. Rabby Wallet’s transaction simulation can help users understand the cost difference between using an official bridge (slower, cheaper in some cases) versus a liquidity pool swap (faster, but with price impact). The key is understanding that “moving to a cheaper network” is itself a transaction with costs, not a free optimization.

Future changes and protocol upgrades affecting fees

Ethereum’s fee structure will continue to evolve. Proto-danksharding (EIP-4844), already live on the Dencun upgrade, uses temporary blobs of data that expire after roughly 18 days. This allows Layer 2s to post data more cheaply because blob storage is subsidized differently than regular calldata. As proto-danksharding becomes more widely used, Layer 2 fees are expected to drop by 5-10x, making transactions cost fractions of a cent rather than cents.

Full danksharding, planned for future upgrades, will increase the number of blobs available per block and further reduce costs. However, these improvements are not magical; they exchange cheaper transactions for slightly increased validator hardware requirements and greater complexity in the protocol. Each optimization trades something for cost reduction, and each trade-off represents a choice about Ethereum’s future direction.

Layer 2s themselves are evolving. Some are experimenting with sequencer auctions that allow external builders to bid for the right to order transactions, potentially creating more competitive fee markets. Others are implementing new compression techniques or different data posting schedules to reduce costs further. The one constant is that fees will remain tied to network capacity, validator economics, and the security model of the underlying chain.

Users who understand these principles can make better decisions about when to pay Ethereum’s high fees for security and finality, when to use Layer 2s for cost and speed, and when to use alternate chains for different trade-offs. To get started managing assets across multiple EVM networks while maintaining full control of your private keys, you can how to download the rabby extension and import or create a wallet. As you become comfortable with different networks, the gas fee differences will become second nature, and you will instinctively choose the network that makes economic sense for each transaction.