If Ethereum mainnet is where the institutional brand sits, the L2 ecosystem is where the institutional payments and high-frequency activity actually runs. The L2 thesis is not "cheaper Ethereum"; it is "Ethereum security inheritance with operational properties (cost, finality, throughput) that make a tokenised payments rail viable." Base is the L2 most relevant to the institutional tokenisation story today, because Coinbase operates it, JPMD launched on it, and the institutional client base that is the natural counterparty for a tokenised deposit token is already deployed there. This part covers why Base specifically, and why the other major L2s (Polygon, Arbitrum, Optimism) matter less for Kinexys today.

What Base actually is

Coinbase's Ethereum L2, built on the OP Stack. Base is an optimistic rollup using Optimism's open-source rollup framework. Coinbase operates the sequencer (the entity that orders transactions before they are batched and posted back to mainnet for settlement). Transactions on Base inherit Ethereum's data availability and ultimate settlement, but execute in a separate, cheaper, faster environment.

The economic profile is the inverse of mainnet. Transaction fees are sub-cent and stable across congestion levels. Block times are roughly two seconds. Throughput is in the thousands of transactions per second range. Withdrawal back to mainnet still incurs the standard seven-day optimistic-rollup challenge period, which matters for the cross-chain liquidity case (covered in Chapter V) but not for activity that stays inside Base.

Why JPMD chose Base

Four reasons, in roughly the order they actually drove the decision.

Coinbase distribution. This is the load-bearing one. Coinbase is the largest US-regulated digital-asset venue with the deepest institutional client base. Those institutions are already deployed on Base, custodying with Coinbase Custody or compatible MPC providers, with Base integrated into their treasury operations. JPMD on Base means JPMD is reachable by that institutional base on day one. JPMD on a chain those institutions are not on means a separate chain integration project before any of them can even hold the token. The distribution argument is structurally similar to MONY's composability argument on mainnet, just translated to a different counterparty universe.

Predictable fees. A payments product processes thousands of transactions daily. On mainnet, gas variance turns operational cost into something between unpredictable and prohibitive. On Base, transaction cost is small and stable. For JPMD, where the product economics depend on the per-transaction cost being negligible and predictable, this is non-negotiable. The same logic applies to corporate treasury use cases where finance teams need to forecast operating costs to the basis point.

Ethereum security inheritance. Base settles to mainnet for finality. The validator set securing Base's ultimate state is Ethereum's validator set. From an institutional risk-diligence perspective, Base is not a separate trust assumption; it is an operational layer above the same trust assumption mainnet provides. That framing matters when an institution's risk committee is asked to approve a new chain integration: "this is Ethereum, with cheaper execution" is a much shorter conversation than "this is a new L1 with its own validator set and its own attack surface."

Institutional credibility. Coinbase is a US-listed public company, regulated under multiple frameworks, with established institutional partnerships. Choosing a Coinbase-operated chain over a more crypto-native L2 signals which side of the institutional/crypto-native line JPM is positioning JPMD on. The signal matters because regulated counterparties watch which chains regulated issuers choose.

What Base costs JPMD

Less crypto-native composability than mainnet. The DeFi ecosystem on Base is real but thinner than on mainnet. For a permissioned payments token, this is largely fine; JPMD is not designed to be wrapped into automated DeFi strategies, and the lack of crypto-native composability is closer to a feature than a bug for a regulated product.

Sequencer centralisation. Base's sequencer is operated by Coinbase. There are roadmaps to decentralise this, common across the OP Stack ecosystem, but as deployed today the sequencer is a single operational point. For an institutional issuer whose own corporate posture is conservative on operational risk, partnering with Coinbase as sequencer operator is a credible answer; partnering with a more diffuse operator would be a harder sell.

Slower mainnet withdrawal. The optimistic-rollup challenge period (seven days) means moving JPMD off Base back to mainnet is slow. For most JPMD use cases this is irrelevant (the token lives on Base). For cross-chain orchestration (e.g., bridging JPMD-equivalent value between Base and a non-OP-Stack chain) this is a constraint that the cross-chain layer (Chapter V) has to handle separately.

Why Polygon, Arbitrum, Optimism matter less for Kinexys today

Polygon CDK is being used by some institutional tokenisation projects, including parts of BlackRock's BUIDL extension across chains and some of the Hong Kong Project Ensemble work. It is a credible institutional L2 in its own right. For Kinexys, the absence of a Coinbase-equivalent distribution argument means there is no compelling reason to add Polygon as a primary chain.

Arbitrum hosts a deep pool of crypto-native institutional activity (market makers, derivatives venues, on-chain treasuries), but the institutional MMF and tokenised-deposit segments have largely settled on mainnet and Base. Adding Arbitrum would replicate Base's economic profile without adding distribution that JPM does not already reach.

Optimism shares the OP Stack with Base. The two chains are technologically near-identical; the difference is who operates the sequencer and who the institutional counterparty base is. JPMD already has Base; Optimism would add operational complexity without adding institutional reach.

The honest framing is that L2 selection for an institutional tokenised product is driven primarily by where the counterparty base sits, secondarily by the chain operator's institutional credibility, and only thirdly by the technical properties (which are largely interchangeable across the OP Stack). Base wins on all three for JPMD today, and adding more L2s without a specific counterparty trigger is operational cost without strategic gain.