Found while measuring the first calibrated rowwise UK local candidate (#762, receipts R8–R15 on uk-rowwise-candidate-762-b). Both rows are reviewed measure exclusions under #762 A16, expiring 2026-10-03, tracked on #736.
Measured on spine-m (FRS 2024-25, policyengine-uk 2.92.1, calibration year 2025), single-block engine resolution:
| row |
spine |
SLC |
miss |
slc.borrowers.plan_2_liable |
3.37M |
9.71M |
−65 % |
slc.borrowers.plan_2_above_threshold |
1.37M |
4.46M |
−69 % |
Block-insensitive, so the misses are real. Under the ruled doctrine the solver moves them to −52 % / −43 % and no further within the stretch bound; the slc.repayments rows fit (England total within 10 % under grain_equal), which points at the stock being wrong while repayment flows are roughly right.
Likely cause: the FRS student-loan flag captures people currently repaying (or reporting a deduction), not the liable stock of plan-2 borrowers — those below the repayment threshold, abroad, or in deferment carry no FRS signal. The SLC statistics publish the liable stock and the above-threshold count directly.
Ask: impute plan-2 borrower status (liable / above threshold) from the SLC borrower statistics — by age cohort and graduate status — as a spine stage, keeping the FRS repayment amount where reported; and declare the reconciliation between the FRS "repaying" signal and the SLC "liable" concept on the target side if the contract means the latter.
Acceptance: both rows within 25 % at initialization on the next spine and the two A16 exclusions retire.
Refs #762, #736, #665.
Found while measuring the first calibrated rowwise UK local candidate (#762, receipts R8–R15 on
uk-rowwise-candidate-762-b). Both rows are reviewed measure exclusions under #762 A16, expiring 2026-10-03, tracked on #736.Measured on spine-m (FRS 2024-25, policyengine-uk 2.92.1, calibration year 2025), single-block engine resolution:
slc.borrowers.plan_2_liableslc.borrowers.plan_2_above_thresholdBlock-insensitive, so the misses are real. Under the ruled doctrine the solver moves them to −52 % / −43 % and no further within the stretch bound; the
slc.repaymentsrows fit (England total within 10 % undergrain_equal), which points at the stock being wrong while repayment flows are roughly right.Likely cause: the FRS student-loan flag captures people currently repaying (or reporting a deduction), not the liable stock of plan-2 borrowers — those below the repayment threshold, abroad, or in deferment carry no FRS signal. The SLC statistics publish the liable stock and the above-threshold count directly.
Ask: impute plan-2 borrower status (liable / above threshold) from the SLC borrower statistics — by age cohort and graduate status — as a spine stage, keeping the FRS repayment amount where reported; and declare the reconciliation between the FRS "repaying" signal and the SLC "liable" concept on the target side if the contract means the latter.
Acceptance: both rows within 25 % at initialization on the next spine and the two A16 exclusions retire.
Refs #762, #736, #665.