Repository navigation
Update the PIP point system docs page (#1069) - #1725
Conversation
463a466 to
56cbca0
Compare
#1069 references SI 2013/377 Part 2 (daily living activities and points) and asks for the underlying point system to be modelled. Currently PolicyEngine UK reads the category (STANDARD / ENHANCED / NONE) directly via pip_dl_category and pip_m_category and multiplies by the weekly rate. That's correct for population aggregates because the FRS doesn't capture activity-level scores and DWP's caseload publications report bands rather than scores. Where the points machinery would matter: descriptor reforms (e.g. the 2024 Green Paper proposals), threshold reforms (raising the 8/12 cutoffs), and any analysis distributing by activity. Three-phase plan: 1. Points-to-category mapping behind a use_point_system switch parameter, preserving the categorical default (10 daily-living activity scores + 2 mobility activity scores -> total -> band via threshold params). 2. Descriptor-level parameters under gov/dwp/pip/descriptors/ so descriptor point values themselves are reformable. 3. Data-side descriptor imputation in policyengine-uk-data (needed to make Phase 2 reforms work on the calibrated microsim). Pure planning; current pip variables and parameters unchanged.
…gelog fragment Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
56cbca0 to
ceedee1
Compare
policyengine.py 6.2.6 certifies policyengine-uk-data 1.58.0 as the UK default; Microcosm UK is the replacement being built. Both derive the fields this page relies on in the same way. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main already carries a later version of this page (added by #2386 and revised on its branch on 19 October); the conflict resolves to main's version before the page is corrected. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- #1661 merged on 21 October 2026: pip_dl_points and pip_m_points and the gov.dwp.pip.{daily_living,mobility}.points_thresholds parameters exist, so the page is "partly planned"; the category comes from points only when none is supplied for the year, and an earlier category carries forward unless points are newer (utils/pip.py). - No data path supplies points: microcosm#145 closed on 16 October with the PIP points item dropped. - Microcosm UK is the certified default (policyengine.py 6.10.0), not a dataset "being built"; both derive bands from reported amounts. - Threshold reforms now work for households entered with points, not in microsimulation; Phase 1's threshold parameters are done. - Activity 8 is "reading and understanding signs, symbols and words" (SI 2013/377 Sch. 1 Part 2). - No DWP table is called "activities most commonly scoring above zero"; the per-activity score distributions are Tables 2.27 and 2.28 of the Pathways to Work evidence pack, chapter 2. - Replaced the released .added fragment with a .changed fragment. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From the independent review of the corrections (household runs on main): - A category supplied for a year is stored for the whole simulation, with NONE for anyone not given one; points count only up to the year being calculated, and an earlier category carries forward unless the points are for a later year than it (utils/pip.py). - DWP's regular statistics report award levels; per-activity figures are one-off, which the Phase 3 citation now matches. Tables 2.27 and 2.28 cover working-age, normal-rules daily living awards in England, Wales and abroad. - "partly modelled" heading; descriptor combinations are drawn within the observed band; #1661's points inputs default to -1 ("not assessed"). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…able
Second-round review (approved): utils/pip.py ignores -1 ("not assessed")
points; the evidence pack has mobility total-points tables but none per
activity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
UK hub (Claude build session, for Max) Head: Context. A later version of this page reached main through #2386 (merged 21 October 2026), and the Is the plan still worth having? Yes. #1661 merged on 21 October 2026 and implements total points and the thresholds. The page now says "partly modelled" and keeps the per-activity scores, descriptor parameters and data imputation as planned work. Review. I had a verifier extract and check every claim, checked each correction myself, and then ran independent adversarial reviews (household runs on main). Round 1 asked for changes, and round 2 approved; its two optional wording points are applied in the last commit. Corrected (13) What #1661 changed
Data and sources Checked and correct (unchanged)
|
Corrects the PIP point system planning page (
docs/book/programs/gov/dwp/pip-points-plan.md), written for #1069.The page itself is already on main: it shipped in #2386 (merged 21 October 2026), and its original
.addedchangelog entry was released in 2.143.0. This branch has merged main, dropped the released fragment and now carries only corrections to main's copy of the page.What this PR corrects (from its changelog fragment):
The claim-by-claim review behind these corrections, with sources, is in the "Ready for hub review" comment on this PR.
changelog.d/1069-pip-points-page-corrections.changed.md(changed).🤖 Generated with Claude Code