This site is an education sandbox: identifiers, hashes, prices and balances are simulated. But the rules, settlement mechanics, costs and precedents it illustrates are real, and every one is listed below with what it's based on and a link to the authoritative source. We prioritise Canadian sources (FINTRAC, Payments Canada, Bank of Canada, CSA, OSFI, Department of Finance); where a figure is a global industry benchmark, it is labelled as such, so nothing is a hidden assumption.
The savings figures shown on the What It Saves page run through a tightened v2 model. This appendix is the audit trail: why v2 exists, every constant, the removal factors that drive 4orm-side savings, and three worked examples spanning a Tier 1 bank, a regional credit union and a community credit union.
Why v2 tightens the model. The v1 model used aggressive removal factors (95 to 100% across the board) and a single national average for transaction volume. v2 separates the four legacy cost categories, applies a different removal factor to each based on what 4orm actually changes, and uses tier-specific transaction volumes drawn from Payments Canada flow data and OSFI asset bands.
The result is more conservative than v1 and easier to defend in a real diligence conversation: every input is a constant you can change, every removal factor has a one-sentence reason, and the worked examples below show the math end to end for three institution types.
| Constant | Value | Source / basis |
|---|---|---|
| Loaded labour rate | C$60 / hour | Canadian AML-analyst median + ~30% overhead load (benefits, supervision, tools). |
| AML alert rate | 0.6 to 1.4% of transactions | Range observed across Canadian Tier 1 and credit-union deployments; tighter monitoring sits at the high end. |
| Minutes per alert (manual review) | 5 to 10 min | First-line analyst triage; complex SAR cases excluded from this range. |
| Reconciliation break rate | 3 to 6% of cross-bank capital-markets / securities trades | Scoped to capital-markets cross-bank settlement (the target of 4orm's atomic DvP). ACSS / Lynx retail-payment break rates are materially lower (typically <1%). |
| Cost per recon break | ~C$120 | Analyst time + ops escalation; based on typical Canadian back-office staffing. |
| ACSS settlement window | 1.5 to 2 business days | Deferred-net cycle; high-value through Lynx settles faster but trapped balances persist overnight. |
| Correspondent fee | C$0.15 to C$0.25 per cross-bank transfer | Net of revenue sharing; varies by counterparty bank and volume band. |
| Cost of funds (trapped liquidity) | 3.5% | Bank of Canada policy rate (2.25%) + ~125 bp blended funding spread; institution-adjustable. |
These are the percentages applied to each legacy cost category to estimate the 4orm-side run-rate. Lower than v1 across the board, and tied to a specific mechanism rather than a single across-the-board number.
The same v2 model run against three Canadian institution profiles. Annual numbers, CAD, illustrative. Institution names below are used solely as size proxies for the model; no endorsement or pilot relationship is implied.
How to update these benchmarks. Every constant above is intended to be revisited with each pilot cohort. Where an institution has its own internal figures (their actual alert rate, their actual recon break cost, their actual cost-of-funds), those replace the defaults on a per-engagement basis. If a published Canadian source moves materially (Bank of Canada rate, FINTRAC fee schedule, ACSS settlement window), this page is updated and the change is noted in the v2 audit log. The goal is not to claim precision the model does not have; it is to show every lever a credit-union CFO or bank treasurer would want to pull when checking the work.