Published on Sat Jul 25 2026 00:00:00 GMT+0000 (Coordinated Universal Time) by Jacob Cavazos

Notional Pooling Meets Blockchain: The Treasury Play

Building on a centuries-old mechanism with a fundamentally cheaper rail

Notional pooling is one of the most elegant tricks in corporate treasury. A multinational’s subsidiaries each maintain separate bank accounts across jurisdictions, but the bank aggregates all balances into a single notional position. No funds physically move. Surpluses in the German entity offset deficits in the Brazilian entity, and the bank calculates interest on the net. The result: reduced external borrowing, maximized interest income, and zero intercompany loan paperwork for the netting itself.

The catch? The entire mechanism depends on bank infrastructure — banking hours, correspondent relationships, cross-guarantee agreements, jurisdiction-specific set-off laws, and FX spreads that typically run 50–150 basis points. It works, but it’s slow, opaque, and the institutions that profit from the friction are the ones providing the service.

Blockchain doesn’t replace notional pooling. It replaces the infrastructure layer that makes it expensive.

Every limitation of traditional pooling — settlement delays, banking-hour dependencies, multi-currency conversion spreads, cross-jurisdictional friction — maps to something that on-chain settlement does fundamentally better. The question isn’t whether companies will move treasury operations on-chain. According to EY-Parthenon’s 2025 survey, 58% of corporates already plan to adopt stablecoins within two years. The question is how they’ll structure it, and who builds the rail.

We’ve been building that rail. Here’s what we’ve learned.

The execution proof

Orkid operates as a UniswapX solver and RFQ aggregator across Ethereum, Base, Unichain, Arbitrum, and Polygon. We fill Dutch auction orders, aggregate quotes from private market makers (Bebop, Liquorice), and execute MEV-protected multi-hop swaps. The same engine handles treasury settlement: large-notional stablecoin movements, cross-chain rebalancing, and intercompany transfers settle in a single block.

As of July 2026, we’ve executed ~31,000 verified transactions across 6 networks:

Every transaction is verifiable on public block explorers. The wallets hold near-zero balances because they’re pass-through executors — funds flow through, they don’t sit. That’s the expected state for infrastructure that routes value.

What changes when settlement costs ~1 bps

Here’s the structural shift that makes on-chain treasury compelling.

Banks charge 50–150 basis points for FX and cross-border settlement. That’s the price the customer pays — it includes the bank’s margin, correspondent banking fees, and infrastructure costs.

On-chain execution costs roughly 1 basis point. That’s gas + calldata on Base or Ethereum. It doesn’t scale with trade size — a $100 swap costs the same in gas as a $100M swap. This is the structural advantage of calldata-based pricing versus notional-based pricing.

We haven’t set customer pricing yet. But the math is straightforward: if execution costs ~1 bps and we charge 5–10 bps, we’re still 5–15x cheaper than bank FX. The margin window between our cost and their price is where the business exists. Whether those savings flow to the customer, to us, or get shared is a business model decision — the cost advantage is what makes it possible.

How on-chain settlement extends notional pooling

Traditional notional pooling relies on the bank’s ledger to net positions. On-chain, a master smart contract reads all subsidiary wallet balances in real-time and computes the net notional position automatically. No reconciliation, no T+1 reporting lag, no waiting for the Singapore office to submit their balance.

When a subsidiary needs to pay a supplier, the master contract can authorize payment from another entity’s surplus wallet. The accounting trail is immutable — every movement is on-chain and auditable. The blockchain doesn’t eliminate intercompany loan documentation, but it makes it radically more transparent and faster to reconcile.

Multi-currency pooling gets particularly interesting. The traditional “Dutch” variant requires the bank to negotiate FX spreads on net positions. On-chain, stablecoin FX routes through automated market makers:

• Subsidiary holds EURC (euro-pegged stablecoin on Base)
• Settlement routes through Uniswap V3 or Curve to convert to USDC at transparent, market-determined rates
• Execution cost: ~1–5 bps on-chain vs 50–150 bps traditional bank FX margins
• Settlement: sub-30 seconds, 24/7/365 — no banking-hour cut-offs

The mechanism is different — AMMs versus dealer desks — but the outcome for the customer is the same currency conversion at a fraction of the cost.

What on-chain settlement does faster, cheaper, and more transparently

Settlement speed. T+1 or T+2 becomes ~200ms on Base, ~12 seconds on Ethereum. For liquidity management across subsidiaries, this collapses the settlement window from days to seconds.

24/7/365 availability. Banks don’t process wires on weekends. On-chain settlement doesn’t have banking hours. Emergency liquidity at 2 AM Saturday is possible — the rail is always open.

Price transparency. Bank FX spreads are negotiated, opaque, and relationship-dependent. On-chain execution prices are visible on public block explorers. “Here’s the proof of execution at 1 bps” is a different conversation than “our bank gave us a good rate.”

Cost at scale. Bank FX scales with notional. On-chain execution is flat — calldata is calldata regardless of how much value it moves. The bigger the trade, the bigger the advantage.

Where the yield layer gets interesting

In the legacy world, the net positive pool balance sits in a bank account earning roughly 3.7% (short-term Treasury equivalent). The bank pockets the spread between what they pay the corporation and what they earn deploying those funds.

On-chain, the net positive balance can be programmatically deployed into yield-generating protocols — automatically, transparently, and with instant recall when liquidity is needed:

The yield premium over T-Bills reflects real risk — smart contract exposure, oracle dependency, and the absence of regulatory recourse. That’s not “free money recovered from banks.” It’s compensation for taking on risks that banks traditionally absorbed. But for capital that’s already on-chain, these are audited, battle-tested protocols with billions in TVL. The operational advantage: when a subsidiary needs liquidity, the master contract can withdraw from Aave V3 in a single block (~12 seconds on Ethereum, ~2 seconds on Base). Traditional banking can’t process a weekend wire at all.

Strategically, Orkid’s captured execution fees are designed to sweep into a Hedera-native staking reserve — turning routing revenue into productive TVL. The execution layer self-funds a yield product. That’s the long-term play: the rail generates the fees, the fees become the reserve, the reserve earns the yield.

A practical path to adoption

Tier 1: Conservative entry. Hold stablecoins in a multi-signature wallet structure across subsidiaries. Use a master contract for balance visibility and intercompany settlement. Deploy idle balances into tokenized T-Bill products (BUIDL at ~3.7%) where the underlying is government debt — familiar, regulated, audited. Settlement time collapses from 3–5 days to seconds. The operational savings alone justify the migration.

Tier 2: Active yield management. Deploy net positive pool balances into Aave V3 and Morpho vaults. The master contract monitors borrowing demand and rebalances between protocols to chase the best risk-adjusted yield. Aave’s $40B+ TVL provides liquidity depth that even a $50M corporate position can enter and exit without meaningful market impact. Yield uplift: +80 to +130 bps over Treasury yields — real money that currently flows to the bank as spread.

Tier 3: Programmable treasury. The master contract becomes a treasury execution engine. It monitors subsidiary balances in real-time, predicts liquidity needs, and automatically deploys surplus capital into the highest-yielding risk-appropriate venue. When a subsidiary’s balance drops below a threshold, the contract recalls funds from lending protocols before a shortfall occurs. This is where Orkid’s execution infrastructure and the reserve layer combine into something that transcends both traditional pooling and standalone DeFi yield.

What this doesn’t replace

We should be clear about what on-chain settlement doesn’t do, because overclaiming is how you lose credibility with the people who actually run treasury operations.

Set-off rights — the legal mechanism that makes notional pooling possible — are jurisdiction-specific and bank-provided. A smart contract aggregating balances doesn’t create set-off rights under German or Brazilian law. The legal architecture stays with corporate counsel. What blockchain does is make the infrastructure cheaper and faster; the legal framework still matters.

Approval workflows — signatory thresholds, SOX controls, board resolutions — don’t disappear because the rail is faster. A multi-sig still needs M-of-N signers. The settlement is instant, but the decision to settle still goes through institutional process.

Regulatory reporting — CTR thresholds, SAR triggers, transfer pricing documentation — become more urgent with real-time settlement, not less. Faster movement means compliance needs to keep pace.

The honest framing: on-chain settlement replaces the infrastructure layer — the rail, the FX conversion, the settlement speed, the audit trail. It doesn’t replace the institutional layer — the legal agreements, the approval workflows, the regulatory compliance. But by making the rail dramatically cheaper and faster, it changes the economics of the entire operation.

Where Orkid fits

We’re building the execution layer that makes on-chain treasury operations cost-effective.

Today: UniswapX solver fills across 5 chains (Ethereum live, 4 chains pending filler contract deployment). RFQ aggregation from Bebop and Liquorice for large-block execution. MEV-protected multi-hop swaps via the OrkidRouter at calldata-based pricing. ~200ms settlement on Base with 98%+ success rate. ~31,000 verified transactions across 6 networks.

In progress: Per-chain filler contract deployment (Base and Unichain contracts deployed; Arbitrum and Polygon pending). Cross-chain settlement rail with Hedera native staking as the reserve layer. Treasury execution API for institutional clients. ZK-blinded cross-chain intents for privacy-preserving settlement.

The strategic moat: Captured execution fees sweep into a Hedera-native staking reserve. Routing revenue becomes productive TVL. The execution layer self-funds a yield product. The more volume flows through Orkid, the more the reserve grows, and the more competitive the yield offering becomes. It’s a flywheel — but one that only works if the execution layer is proven, which is why those 31,000 transactions matter.

For banks specifically

The strategic question for banks isn’t whether to adopt on-chain settlement — it’s whether to build it themselves and capture the infrastructure fees, or partner with providers who already have it running.

58% of corporates plan to adopt stablecoins within two years. Those corporates will need settlement infrastructure. Banks can be the ones offering it — using infrastructure like Orkid’s — or they can watch fintechs and DeFi protocols capture the treasury services revenue one basis point at a time.

The technology exists. The protocols are audited. The regulatory framework (GENIUS Act) is in place. The remaining barrier is institutional inertia — the comfort of doing what’s always been done, even when what’s always been done costs more, settles slower, and earns less.

For banks that want to be the provider rather than the displaced: we should talk.

All transaction counts verified via public block explorers (Basescan, Etherscan, Arbiscan, Polygonscan, Uniscan) and Hedera HashScan on July 13, 2026. Wallet addresses and transaction histories are publicly auditable. For integration inquiries, contact Orkid Labs.

Written by Jacob Cavazos