Repository navigation
Count Carer Support Payment after the overlapping-benefit reduction in Scottish working-age Council Tax Reduction - #2821
Conversation
Scottish Council Tax Reduction income counted carer_support_payment_pre_overlap, so a carer whose State Pension or contributory ESA reduces Carer Support Payment to nil under SSI 2023/302 reg 16(2) still had 86.45 a week counted. SSI 2021/249 reg 57(1)(b) counts the CSP the carer is entitled to be given, and Social Security Scotland's guidance reduces CTR by the CSP received; the pension-age SSI 2012/319 reg 27(3) text leaves both readings open, so the agency's reading applies at both ages. Fixes #2820 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Aggregate impact on certified data: passBase Checks: 0 broken by this change, 1 failing on both sides, 0 fixed. Moves flagged for review: 0 (thresholds in Notes
Moves flagged for reviewNone. Spending by benefit and service
Caseloads
Household net income by decileNo household's net income changes in any year. No change in any year: receipts by tax head; government balance, each component counted once; model totals; household income totals; population; poverty rates; people in poverty; inequality; mean household net income in each side's own deciles; standard reform costings (change in gov_balance); standard reforms: change in each reform's target. Checks failing on both base and head (1)
Counted-once checkThe check raised each of 92 components by £1 per unit in 2026-27 and read every head of its totals household by household: its own head must rise by exactly £1, no other tax head may move, no other benefit may rise, and the model's own total must move by what its heads moved. A response that is not finite breaks every rule. The check measures a head that breaks a rule twice more: once holding every other head at its baseline value, and once holding every other head at its value after the first £1 while the component rises again. The head fails where either measurement still shows the response. The check cannot hold a head at a value that is not finite. In each unit where another head is not finite after the first £1, it holds that head at its baseline value, and the second measurement judges no household the unit belongs to. There a head fails wherever it first broke a rule, or wherever the check could not confirm whether its response contains the component. 0 failed and the check skipped 0.
Provenance
Dataset: Receipts by tax head: head levels
Spending by benefit and service: head levels
Caseloads: head levels
Government balance, each component counted once: head levels
Model totals: head levels
Household income totals: head levels
Population: head levels
Poverty rates: head levels
People in poverty: head levels
Inequality: head levels
Mean household net income in each side's own deciles: head levels
Standard reform costings (change in gov_balance): head levels
Standard reforms: change in each reform's target: head levels
|
juaristi22
left a comment
juaristi22
left a comment
There was a problem hiding this comment.
Reviewed adac4407925c860e23ac5ae21174052a7f4ee8a9. PARTIAL: no confirmed defects; one material interpretation gap remains.
The code checks passed: 63 tests and three final CTR/household-income diagnostics, including nil and partial CSP overlap. Underlying entitlement and the carer premium are retained. Current-head required checks also pass.
Before I can approve the pension-age change, I need a CTR-specific explanation of how SSI 2012/319 regulation 27(3) and (5) interacts with CSP regulation 16(2). Regulation 27 normally counts benefit before deductions, and its exceptions do not expressly name CSP regulation 16. The competing reading is that regulation 16 fixes the amount initially payable rather than deducting from a payment. This distinction controls the changed pension test (£799.14 difference).
The general Social Security Scotland guidance supports counting CSP received, but does not reconcile those pension-CTR provisions. Housing Benefit regulation 29 expressly names the CSP adjustment, so the HB-equality test cannot settle CTR's interpretation. I am not asserting that the chosen paid-CSP result is legally wrong.
I checked the actual Axiom #432 body and posted output-selection comment; that follow-up is recorded correctly. No billed encoding or private population run was performed.
Please add authority or an authoritative administrative interpretation that directly addresses the regulation 27/regulation 16 distinction. Carer-premium retention is independently supported. The PR also remains a draft, so this is a review comment rather than an approval.
|
Thanks for the review, @juaristi22. We searched the web, Find Case Law and the instruments' published notes and found no decision or guidance that reconciles reg 27(3) with reg 16(2). BAILII's bot check stopped our search there, and the Local Taxation Chamber, which hears CTR appeals, doesn't publish its CTR decisions. The Scottish Courts and Tribunals Service lists 146 Upper Tribunal for Scotland decisions on appeals from that Chamber and the Social Security Chamber, and none cites reg 27(3) or reg 16(2). We haven't asked the Scottish Government how it reads the two together. So we've made a judgment call: on the words alone we'd count CSP before the cut, on the Scottish Government's stated purpose we'd count it after, and we follow the purpose. The council counts a benefit before "any deduction" (reg 27(3)), and the Scottish Ministers listed no CSP cut among the exceptions (reg 27(5)). They wrote "reduced" in reg 16(2) and kept "deduction" for setting CSP against a liability (reg 26). The Scottish Government's Policy Note for the CSP Regs says "the amount of Carer Support Payment may be reduced or set to £0". The Scottish Government told the Scottish Commission on Social Security that a carer on a £0 award "remains entitled to Carer Support Payment but does not receive payments". That supports the line you drew between setting CSP's amount and deducting from a payment. But in reg 27(5)(c), the Ministers carved "reductions in incapacity benefit" out of "any deduction", so "any deduction" in reg 27(3) can cover a reduction. They also named reg 16 beside the 1979 Regulations in the carer premium rules (Sch 1 para 6(1)(ab)). In SSI 2025/340 Sch 1 para 8, they amended reg 27(1)(j) and that reference without adding reg 16(2) to reg 27(5). The UK Government added it to the matching Pension Credit, pension-age Housing Benefit and English pension-age CTR rules. The Scottish Ministers amended both CTR schemes for CSP in SSI 2023/258, and the Scottish Government's Policy Note for that instrument comes closest to the question. It says the amendments treat CSP payments and clients "consistently with payments and clients of Carer's Allowance". It adds that they make councils take CSP "into account in assessing entitlement" to CTR and let carers "receive additional amounts" there "in recognition of their caring role" (paragraph 9). It gives the instrument's purpose as ensuring that "carers receiving Carer Support Payment are not disadvantaged compared to carers receiving Carer's Allowance" (paragraph 5). The note doesn't mention reg 27, reg 16 or overlapping benefits, and the Ministers left reg 27 alone in that instrument. Because the Ministers listed the 1979 Regulations in reg 27(5)(a), the council counts none of the Carer's Allowance that DWP cuts for a carer's State Pension. A council that counted CSP before the cut would count both the State Pension and the CSP it replaces. For a carer on a nil award in 2026, it would also add £86.45 a week of income against a carer premium of £48.15, so a claim that pays nothing could cost the carer CTR. For example, the couple in our pension-age case gets £1,599.18 of CTR a year if the partner doesn't claim CSP ( The Reading section in the PR description has the full reading with links. If you read the words as decisive, we'd switch pension-age CTR to |
|
Hand-off to the UK hub (
|
juaristi22
left a comment
juaristi22
left a comment
There was a problem hiding this comment.
Source-only follow-up at unchanged adac4407925c860e23ac5ae21174052a7f4ee8a9 — PARTIAL; no confirmed findings, one material legal-interaction gap remains.
The new official Policy Notes support the intended Carer’s Allowance parity/net-payment treatment and explain that regulation 16(2) reduces the CSP amount. They do not expressly reconcile that reduction with pension-age CTR regulation 27(3)/(5), its broad pre-deduction wording, or the absence of the express exception found in the Housing Benefit comparator. The remaining gap is this statutory interaction, rather than the unsuccessful search for a published decision.
The new argument narrows the question but does not establish an unconditional wrong-cash oracle for either reading. Prior verified code evidence (63 repository checks and three native cases) is retained as historical evidence; no model tests or calculations were repeated. Approval remains withheld pending evidence resolving the interaction.
Official sources: SSI2023/258 Policy Note, SSI2023/302 Policy Note.
Main's devolved working-age route (#1966) gave Scottish working-age claims their own unearned income, which counted Carer Support Payment before the overlapping-benefit reduction. This branch counts it after the reduction at working and pension age, so the merge carries that reading into council_tax_reduction_working_age_unearned_income too, with the SSI 2023/302 reg 16 and Social Security Scotland references. The contributory ESA case now expects nil CSP on the working-age route as well (working-age applicable income 5,000 and an award of the whole 2,000 bill), and invariant 5 of the carer income properties compares the rise in working-age unearned income with the CSP paid after the reduction. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
We merged main (e050200) into this branch. Main's working-age route (#1966) counted Carer Support Payment before the overlapping-benefit reduction, so in the merge we switched |
juaristi22
left a comment
juaristi22
left a comment
There was a problem hiding this comment.
Follow-up at d5a84f14c17140e99d9691f78c8841b33ad4d16c — PARTIAL: no confirmed findings; the pension-age legal-interaction gap remains unverified.
I rechecked the merged working-age route, the changed tests and affected dependencies. 100 repository tests passed (96 YAML cases and four Python property tests), with no failures or skips, plus four household-context evaluations through final CTR and household net income. The 2026 contributory-ESA example counts £5,000 of income and awards £2,000 CTR; restoring the captured-main income readers gives £1,595.40 CTR, a £404.60 final-income difference.
The working-age post-overlap reading is supported by SSI 2021/249 regulations 4 and 57, read with CSP regulation 16(2). The overlap carer premium remains available subject to its other qualifying conditions.
At pension age, the remaining question is whether regulation 16(2) sets the initial CSP amount or is a deduction caught by SSI 2012/319 regulation 27(3), without a listed exception in 27(5). The SSI 2023/258 Policy Note, page 2 supports the parity purpose but does not reconcile that interaction. The pension example's £799.14 CTR/income difference demonstrates the implementation; it does not establish which legal reading is correct.
The current RuleSpec #432 packet and both comments were checked. Approval remains withheld pending authoritative evidence resolving the pension-age interaction or a scope change. Exact-head CI was still pending in the report capture; local test passes are separate evidence.
Scotland's pension-age scheme keeps main's reading. SSI 2012/319 reg 27(3) counts a benefit before any deduction other than the reg 27(5) adjustments, and reg 27(5) doesn't list the CSP Regs reg 16(2), though the Pension Credit and pension-age Housing Benefit regulations do (SI 2002/1792 reg 15(4)(g); SI 2006/214 reg 29(4)(g)). Until a ruling settles that question, the shared formula, which Scottish pension-age claims follow, counts the CSP before the reduction. The devolved working-age route keeps the after-reduction reading. Invariant 5 now states the sign rule: on the devolved route a claim that brings only the carer premium can raise the award, and a working-age ESA overlap example pins it. The ESA YAML case drops the shared-formula outputs, which a Scottish working-age award doesn't read. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
changed the title
|
Thanks, María. I've narrowed the PR to working age at 445f873. Pension-age claims keep main's reading, which counts CSP before the reg 16(2) cut, so #2820 stays open for them until a ruling settles reg 27(3) and (5). Here's what changed since d5a84f1:
|
juaristi22
left a comment
juaristi22
left a comment
There was a problem hiding this comment.
Follow-up at 445f873074303467e5a979ee4d80fbe1dd56115b — COMPLETE: no findings or material evidence gaps in the narrowed working-age scope. Approval is withheld while exact-head CI remains pending.
The prior pension-age gap is resolved by scope removal. The shared income reader's executable methods match captured main, and the pension example again awards £1,200.86. This preserves main's behavior; it does not settle its pension-age legal interpretation. The description leaves #2820 open for that question.
The working-age post-overlap treatment is supported by SSI 2021/249 regulations 4 and 57, CSP regulation 16(2), and 2018 Act section 24. The overlap carer premium remains subject to its other qualifying conditions.
103 repository tests passed (96 YAML cases and seven Python tests), with no failures or skips. Four final-cash evaluations confirmed working-age CTR/household income of £2,000/£5,000 versus £1,595.40/£4,595.40 under captured-main readers, while pension-age results stayed unchanged. Two separately qualified reported-input controls also passed; their final award was held, so they establish the input-source contract rather than an independently simulated cash effect. Initial reviewer diagnostic setup errors were corrected and excluded, without rerunning the repository suite.
The filed RuleSpec #432 scope addendum selects the working-age change and defers pension-age treatment. The completed code/source review can be reused when CI finishes; approval also requires no new material commits, source changes or review feedback.
juaristi22
left a comment
juaristi22
left a comment
There was a problem hiding this comment.
CI-only approval at 445f873074303467e5a979ee4d80fbe1dd56115b.
The updated description was also checked against exact-head CI logs: 6,087 YAML passes and 12,784 Python passes, with 61 allowlisted skips and two expected failures accounted for by Test / Result.
The completed working-age review remains COMPLETE, with no findings or material evidence gaps. All 20 latest checks now pass at this exact head, including required Lint and Test / Result. Head, description, human feedback and source guards were refreshed; the previous 103-test code/source review is retained without rerunning it.
The updated Aggregate report is attributable to this head and introduces no material contradiction. Its certified-dataset, year, suppression/alignment and unchanged baseline-check limitations remain; no new population score is independently certified here.
The pension-age issue remains outside this narrowed change, with main's behavior preserved rather than its legal interpretation certified. Prior finding and scope dispositions remain unchanged.
Addresses #2820
Scottish working-age Council Tax Reduction (CTR) now counts the Carer Support Payment (CSP) after the overlapping-benefit reduction in the Carer Support Payment Regulations (SSI 2023/302) reg 16(2). The working-age route counted
carer_support_payment_pre_overlap, which added £86.45 a week in 2026-27 for a carer whose contributory ESA or another overlapping benefit reduces CSP to nil. That carer's CTR income now leaves CSP out, and the carer keeps the carer premium. Pension-age claims keep main's reading, which counts CSP before the reduction, so #2820 stays open for them.Reading
At working age, the council counts the benefits "to which the applicant is entitled, subject to any adjustment to the amount payable in accordance with Regulations under section 73 of the Social Security Administration Act 1992", CSP among them (SSI 2021/249 reg 57(1)(b)(iva)). The Scottish Ministers made the CSP Regs under the Social Security (Scotland) Act 2018, not under s. 73, so the council looks to the carer's entitlement. The Ministers must give a carer whatever assistance a determination entitles them to (2018 Act s. 24). Under reg 16(2) of the CSP Regs, Social Security Scotland cuts the CSP "to be given" to a carer by any overlapping benefit paid, to £0 where the benefit matches or exceeds it. The Ministers also defined CSP in reg 4(1) as the component "payable under" the CSP Regs. So the council counts the CSP after the cut.
The Scottish Government calls a carer on a nil award "entitled to an award of £0" (response to the Scottish Commission on Social Security). Social Security Scotland's CSP decision-making guide lists CTR among the benefits that fall by "the amount of Carer Support Payment received". Counting CSP before the cut would count both the ESA and the CSP it replaces.
Pension age
The shared formula,
council_tax_reduction_applicable_income, still counts CSP before the reduction for pension-age claims, as main does. At pension age the council counts social security benefits as income (SSI 2012/319 reg 27(1)(j)). Where a benefit's payment "is subject to any deduction (other than an adjustment specified in paragraph (5))", the council counts "the amount before the deduction is made" (reg 27(3)). In reg 27(5), the Scottish Ministers listed the Social Security (Overlapping Benefits) Regulations 1979 and three other adjustments, but not reg 16(2). The UK Government added reg 16(2) to the matching Pension Credit and pension-age Housing Benefit rules (SPC reg 15(4)(g); HB(SPC) reg 29(4)(g)). The Ministers brought CSP into both CTR schemes with SSI 2023/258, and the Scottish Government's Policy Note says the amendments treat CSP payments and clients "consistently with payments and clients of Carer's Allowance". The note doesn't mention reg 27, reg 16 or overlapping benefits. We found no decision or guidance that reconciles reg 27(3) with reg 16(2), so the formula keeps main's reading until a ruling settles it. The pension-age overlap case in the YAML file keeps main's award of £1,200.86.Changes
council_tax_reduction_working_age_unearned_income, the unearned income on the Scottish and Welsh working-age route (Give Scotland and Wales working-age CTR their own rules #1966), listscarer_support_paymentin place ofcarer_support_payment_pre_overlap. Its references add SSI 2023/302 reg 16, the 2018 Act s. 24 and the Social Security Scotland guide.council_tax_reduction_applicable_incomekeepscarer_support_payment_pre_overlap. Its comment now gives the working-age route's reading and the open pension-age question, and its references add SSI 2023/302 reg 16, SPC reg 15 and HB(SPC) reg 29.council_tax_reduction_carer_support_payment.yaml, a carer on contributory ESA of £96.15 a week, now expects working-age income of 5,000, the ESA alone, and an award of the whole £2,000 bill. Main expects 9,495.40 and £1,595.40.test_council_tax_reduction_carer_income_properties.pychecks the CSP paid on the working-age route and runs the ESA carer as an explicit example.disable_simulated_benefitsreform setscarer_support_paymentto the reported Carer's Allowance for Scottish carers from 2025 and keepscarer_support_payment_pre_overlapsimulated. Under it, the working-age route now counts the reported amount, as Housing Benefit, Pension Credit and Universal Credit do.Invariants
The docstring of
test_council_tax_reduction_carer_income_properties.pylists six invariants for Scottish families with one carer in 2026. We restate invariant 5, on the working-age route:Invariants 1 to 4 and 6 keep main's wording.
Tests
CI passes every check at
445f87307:test_council_tax_reduction_carer_income_properties.py.Impact
On the certified 2024-25 dataset, this change moves no household's net income from 2025-26 to 2031-32 and leaves tax receipts, benefit spending, poverty and inequality unchanged. It breaks none of the Aggregate impact checks.
axiom: TheAxiomFoundation/rulespec-uk#432 queued
Reproducer:
policyengine_uk/tests/policy/baseline/gov/local_authorities/council_tax_reduction/council_tax_reduction_carer_support_payment.yaml::A working-age carer whose contributory ESA reduces Carer Support Payment to nil has none counted in Council Tax Reduction,policyengine_uk/tests/test_council_tax_reduction_carer_income_properties.py::test_council_tax_reduction_income_counts_the_carer_support_payment_component🤖 Generated with Claude Code