# Stage 117 — Resource stress of the monthly supplement payer

## Result

Of **1,404 stressed policy cases**, **1,008 pass all four gates** and **396 fail**. The gates are timely owner-tax collection, timely supplements, both recipients' monthly consumption floors and both annual floors. In this particular grid the four gates have the same pass/fail classification; this is an observed result, not a reason to combine their definitions.

There are 156 matched stressed no-supplement controls. Original recipient references and supplement schedules are held fixed, so a shock cannot make the test easier by lowering the floor.

## Discrete boundary

| Recipient floor | Highest tested income retention that fails | Lowest tested retention that passes |
| --- | ---: | ---: |
| 90% | 4.0% | 4.5% |
| 95% | 5.0% | 5.5% |
| 100% | 6.0% | 6.5% |

These values refer to **owners' frozen outside-income component**, not their total resources, total wealth, profits of all firms or national output. The same tested boundary appears for all three owner-response values and all four opening-cash retentions. No interpolation is performed: a passing point is not an exact critical value, a confidence interval or an empirical parameter estimate.

At the last failing point, with zero retained opening owner cash and q=0.5, the first overdue supplement is in January of year 10 for all three floors. For example, at a 90% floor and 4% retained outside income, the minimum annual retentions are 88.56% and 87.81%, below the declared 90% threshold.

## Declared stress

The opening owner cash stock retains 0%, 25%, 50% or 100% of its source value. Owner outside income is reduced permanently from year 1. The initial coarse grid retained 0%, 5%, …, 100%; after locating failures below 10%, it was refined to 0.5 percentage-point spacing from 0% through 10%. The resulting 39 distinct income values give 4 × 39 × 3 floors × 3 owner q values = 1,404 policy cases.

The numerical refinement was prompted by the coarse results; it is not a prespecified empirical test or statistical coefficient fitting. Every tested passing point is archived rather than assuming monotonicity.

The unchanged stage-116 monthly simulator has zero supplemental treasury seed, no borrowing and no external backfill. Schedules remain the original stage-115 static requirements; recipients do not receive a smaller target when payer resources deteriorate. Each stressed scenario uses a matched control for the incremental owner spending comparison.

## Why the large margin needs qualification

The stress removes only the declared owner outside-income component and opening liquid cash. Personal-interest receipts and government bond-principal redemption are unchanged. Existing ownership, future source income shape and frozen purchase plans are inherited. The market's initial counterparty endowment follows the stressed source income schedule, and is identical between the policy and its matched stressed control; the cross-scenario endowments need not match.

Thus “4.5% retained income passes” does not mean that a real economy could withstand a 95.5% income collapse. Source income is concentrated, some owner cash flows are protected from the shock, recipient income is fixed and there is no production or dividend feedback. The test reveals the arithmetic margin of this specific payer account and also exposes which assumptions require further work.

The identical boundary across opening-cash and owner-response variations is specific to the tested resource levels and timing. It does not prove cash reserves or behavior never matter. Failed cases remain in the archive, with first arrears dates and outstanding balances.

## Validation

Six tests cover isolated shock application and input validation, complete grid and unchanged recipient references, recovery of the unstressed stage-116 cases, cash and arrears identities including failures, exact passing-point extraction and archive reproduction.

The original stage-116 simulator was checked against main and was not modified. All policy paths use its full monthly cash settlement and matching debt claims. The source assumptions, schedule and script hashes are in the archive.

```sh
python scripts/funded_floor_stress_v117.py --output data/processed/funded_floor_stress_v117.json
python scripts/test_funded_floor_stress_v117.py
```

[All stressed cases, controls and boundaries](data/processed/funded_floor_stress_v117.json) include separate gates, receipt/payment balances, below-floor months, first failures and owner spending relative to matched stressed controls. The funding claim still concerns incremental supplements only, not the whole public budget.

## Next research step

Audit the payer's resource base: separate wages, dividends, public interest, private interest and principal receipts, and distinguish recurring taxable income from asset redemption. Then test a levy constrained by a declared recurring-income base and connect redistribution to corporate investment and income feedback. This avoids treating a broad cash-resources test as evidence of a sustainable real-world tax base.