Dudley 2025/26 Council Tax Reduction mismatch against entitledto

[MaxGhenis]

During validation for PR #1534, I found a Dudley mismatch between the public entitledto calculator and Dudley Council's published 2025/26 Council Tax Reduction scheme.

Stroud matched entitledto in a separate spot check, so this does not look like a blanket mismatch across all authorities.

Scenario used in the Dudley spot check:

• postcode DY1 1HF
• local authority Dudley
• working-age single adult
• owner-occupier
• council tax band C
• annual council tax liability GBP 1,359.05
• no children
• no savings
• income-based JSA / claims all entitled benefits

entitledto returned:

• GBP 15.64 per week of Council Tax Reduction
• GBP 10.43 per week of council tax left to pay

PolicyEngine UK and Dudley's published scheme both imply:

• a band C maximum eligible liability cap
• a 40% maximum support rate for working-age households
• GBP 543.62 per year of Council Tax Reduction
• GBP 815.43 per year of council tax left to pay

Published Dudley sources:

• https://www.dudley.gov.uk/residents/benefits/council-tax-reduction-scheme/
• https://www.dudley.gov.uk/media/khllf4zc/dudley-council-tax-reduction-scheme-2025-26.pdf

Relevant points from the Dudley scheme PDF:

• the document states that the working-age Council Tax Reduction scheme is a local scheme determined by the Council, while pensioner rules follow the national prescribed requirements
• paragraph 23 sets a GBP 16,000 capital limit
• paragraph 30A sets a flat GBP 5.00 non-dependant deduction for the capped working-age calculation

Broader legal structure:

• English billing authorities make local working-age CTR schemes under the Local Government Finance Act 1992, Schedule 1A and section 13A
• the Dudley scheme document itself says it is determined by full Council

Open question:

• is entitledto encoding an older Dudley scheme, applying a different liability basis, or making an additional assumption that we are not matching?

If anyone has a current Dudley validation case or direct visibility into entitledto's Dudley rule set, that would help pin down where the difference comes from.

[MaxGhenis]

I double-checked the Dudley numbers against the council's published 2025/26 scheme and the mismatch looks even more like a reversal between the support amount and the residual bill.

Using Dudley's published Band C annual charge of GBP 1,359.05:

• Weekly liability is 1359.05 / 52 = 26.1356.
• Paragraph 29(2) of the scheme says the maximum reduction for a non-pensioner is 40% of the lower of actual liability or Band C liability.
• That gives a maximum weekly CTR of 26.1356 * 0.4 = 10.4542.
• The remaining weekly bill is therefore 26.1356 * 0.6 = 15.6813.

PolicyEngine's annual outputs line up with that arithmetic:

• council_tax_reduction = GBP 543.62/year, which is GBP 10.45/week.
• council_tax_less_benefit = GBP 815.43/year, which is GBP 15.68/week.

The entitledto figures recorded in the original post were:

• GBP 15.64/week CTR
• GBP 10.43/week left to pay

Annualised, those are:

• 15.64 * 52 = GBP 813.28/year
• 10.43 * 52 = GBP 542.36/year

Those annualised values are almost exactly the same split as PolicyEngine's result, but flipped between the CTR amount and the amount left to pay.

That suggests one of two things:

1. the two values were effectively reversed in the entitledto output, or
2. entitledto was applying a 60% maximum-support / 40% minimum-payment interpretation that conflicts with Dudley's published 2025/26 scheme text.

For reference, the Dudley PDF provisions I checked were paragraph 29(2) on the 40% maximum reduction and paragraph 30A on the GBP 5.00 x 1/7 non-dependant deduction.

[MaxGhenis]

I ran the same public-calculator spot checks across entitledto, Turn2us, and the Policy in Practice Better Off Calculator for the same common household:

• single working-age claimant
• owner-occupier
• no children
• no savings
• income-based JSA

Councils checked in that common scenario: East Hertfordshire, Warrington, Dudley, Stroud.

I also ran two additional East Hertfordshire-only entitledto checks:

• lone parent, one child, income-based JSA
• couple, no children, income-based JSA

So the total here is 3 unique household types, but only 1 household type repeated across all 3 calculators and all 4 councils.

Council	Published / encoded expectation	entitledto	PiP Better Off Calculator	Turn2us	Takeaway
East Hertfordshire	91.5% max award for this passported JSA case on Band D	GBP 31.59/week CTS, GBP 2.93/week left on a GBP 34.52/week bill (91.5%)	GBP 137.25/month CTS on GBP 150/month liability (91.5%)	Reached results, but no CTS line appeared	entitledto and PiP agree with the scheme and with the current PE encoding
Warrington	Class D passported JSA case: 91.5% on Bands B to H in the published 2025/26 scheme	GBP 26.69/week CTS, GBP 2.48/week left on a GBP 29.17/week bill (91.5%)	GBP 95.06/month CTS on GBP 126.74/month liability (75%)	Reached results, but no CTS line appeared	PiP diverges here. The PiP amount is roughly Band A sized, and does not match the published scheme text, entitledto, or the current PE encoding
Dudley	40% maximum support for this working-age case, with the published Band C cap	GBP 15.64/week CTS and GBP 10.43/week left to pay, which is effectively the reverse of the published scheme split	GBP 45.30/month CTS on GBP 113.25/month liability (40%)	Reached results, but no CTS line appeared	PiP matches Dudley's published scheme and the current PE encoding. entitledto does not
Stroud	Up to 100% support in this case under the local scheme summary	Full CTS on an annual liability of GBP 1,866.47	GBP 148.09/month CTS on GBP 148.09/month liability (100%)	Reached results, but no CTS line appeared	entitledto and PiP both behave like the full-support scheme

A few takeaways from this sweep:

• Turn2us says its calculator covers Council Tax support in most cases, but on these four authority-specific tests it returned the same JSA / Universal Credit lines for every authority and no CTS line. On these cases it was not usable as an authority-specific CTR comparator.
• entitledto and PiP both look good on East Hertfordshire and Stroud.
• entitledto looks wrong on Dudley for the tested case.
• PiP looks wrong on Warrington for the tested case.
• This is enough to justify treating public calculator comparisons as a harness, not as a single ground truth source.

Additional entitledto-only checks in East Hertfordshire:

• lone parent, one child, income-based JSA: still 91.5%
• couple, no children, income-based JSA: still 91.5%

Those extra household types were not rerun in PiP or Turn2us, so I am treating them as supporting spot checks, not part of the cross-calculator matrix above.

[MaxGhenis]

Follow-up after commit 303beb43 on the draft PR.

I fixed one real PE issue before extending the household matrix: passported income-based JSA / Income Support / income-related ESA cases now stay out of the taper consistently in the currently encoded English working-age schemes, and Warrington's band-cap logic now uses that same passporting condition. The regression file now passes with those cases added.

Confirmed PolicyEngine UK outputs for the four household types now look like this:

Household type	East Hertfordshire	Warrington	Dudley	Stroud
single working-age, income-based JSA	CTR GBP 1,647, left GBP 153	CTR GBP 1,391.61, left GBP 129.27	CTR GBP 543.60, left GBP 815.40	CTR GBP 1,777.08, left GBP 0
couple, income-based JSA	CTR GBP 1,647, left GBP 153	CTR GBP 1,391.61, left GBP 129.27	CTR GBP 543.60, left GBP 815.40	CTR GBP 1,777.08, left GBP 0
lone parent, one child, income-based JSA	CTR GBP 1,647, left GBP 153	CTR GBP 1,391.61, left GBP 129.27	CTR GBP 543.60, left GBP 815.40	CTR GBP 1,777.08, left GBP 0
single pension-age, State Pension + Pension Credit	full liability covered	full liability covered	full liability covered	full liability covered

Those figures now line up with the current encoded scheme expectations for the six modeled authorities, and the new regression cases pass locally.

Additional public-calculator results I have now confirmed:

entitledto in East Hertfordshire:

• couple, no children, income-based JSA: GBP 31.59/week CTR, GBP 2.93/week left on a GBP 34.52/week bill
• lone parent, one child, income-based JSA: GBP 31.59/week CTR, GBP 2.93/week left on a GBP 34.52/week bill

Those two entitledto checks match the 91.5% East Herts structure and the current PE encoding.

Turn2us in East Hertfordshire:

• couple, income-based JSA: results page reached, no CTR line shown
• lone parent, one child, income-based JSA: results page reached, no CTR line shown
• single pension-age, Pension Credit / State Pension: results page reached, no CTR line shown

So Turn2us still does not look usable here as an authority-specific CTR comparator, even once the household shape varies.

I have not yet rerun the expanded household matrix through the public Policy in Practice Better Off Calculator; the earlier PiP checks in this thread are still the current validated PiP spot checks.

[MaxGhenis]

One compact way to score the public calculators is: a household/council combination counts as right if the calculator returns a CTR amount that matches the published scheme structure for that combination. If the calculator returns no CTR line at all, I am counting that as not right for this purpose.

On the common cross-tool matrix (4 councils x 1 common household type = 4 combinations):

• entitledto: 3/4 = 75% right
• Policy in Practice Better Off Calculator: 3/4 = 75% right
• Turn2us: 0/4 = 0% right

Those four common combinations are:

• East Hertfordshire, single working-age, income-based JSA: entitledto right, PiP right, Turn2us no CTR line
• Warrington, single working-age, income-based JSA: entitledto right, PiP wrong, Turn2us no CTR line
• Dudley, single working-age, income-based JSA: entitledto wrong, PiP right, Turn2us no CTR line
• Stroud, single working-age, income-based JSA: entitledto right, PiP right, Turn2us no CTR line

On the wider set of combinations I have actually validated for each tool so far:

• entitledto: 5/6 = 83.3% right
• Policy in Practice Better Off Calculator: 3/4 = 75% right
• Turn2us: 0/7 = 0% right

Those wider denominators are different because I have more extra-household checks in entitledto and Turn2us than in PiP:

• extra entitledto checks in East Hertfordshire: couple on income-based JSA right, lone parent with one child on income-based JSA right
• extra Turn2us checks in East Hertfordshire: couple on income-based JSA no CTR line, lone parent with one child on income-based JSA no CTR line, single pension-age household with State Pension + Pension Credit no CTR line

So the current evidence says:

• entitledto is mostly right on the tested cases, with the main problem being Dudley
• PiP is mostly right on the tested cases, with the main problem being Warrington
• Turn2us has not yet produced an authority-specific CTR result on the tested cases