Repository navigation
US newsletter, 29 October 2026: Prop 40 webinar tomorrow, BLS seminar and what's new since 8 September - #35
Merged
Merged
Conversation
Screenshot of policyengine.org/us/california-wealth-tax on Forbes data as of 2026-10-29: Berkeley assumptions +$103.1B to Hoover assumptions -$34.1B, split by assumption. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… and what's new since 8 September Leads with tomorrow's webinar, "Why the Prop 40 revenue estimates differ", and the calculator's bridge chart on Forbes data as of 29 October. Then the 8 October BLS seminar, the graded 2025 poverty prediction (carried from the unsent 30 September draft), PolicyBench, the JOSS paper, model changes since 8 September, the Axiom Foundation's PBIF cohort place and Max's APPAM panel. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Describe Prop 40 by its residency date (Jan 1), valuation date (Dec 31), rate ramp ($1B-$1.1B) and real-property exclusion. - State the heating and SSI counts as additions since September 8. - Table-cell Register button so classic Outlook keeps its padding. - Senior poverty miss: 0.8 points to between 0.1 and 0.3, per the article. - APPAM session is a roundtable. - Active voice in alt text, headings, Axiom section and footer; full Axiom Foundation name in the contents line. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New section after the Prop 40 box: the six measures on David Trimmer's dashboard (ballot-measures-2026, data manifest 2026-10-28, policyengine-us 2.72.4), each with PolicyEngine's headline figure next to the official estimate and a link to its page under policyengine.org/us/ballot-measures-2026 (served once policyengine-app-v2#1415 merges). Preheader now mentions the dashboard. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Replace the two "In the model" paragraphs with the "Model updates" format of the April 15 and May 19 US editions: a count of policyengine-us pull requests since September 8 (534 merged as of today), five US model bullets linked to the policyengine-us changelog, and four US data bullets from Microcosm (#1044 2025 ASEC pool, #1102 state x AGI-band SOI targets, #1043 119th-Congress district crosswalk, #923 ACS usual hours). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- South Dakota: name the AG/sponsor disagreement, the scenario's dates, the modeled year (2028) and the enrollment basis (DSS monthly average, July 2024 to June 2025). - Proposition NN: add what the state does with retained revenue. - Initiative 645: the statutory bar covers taxes on individual income. - North Dakota: the official range covers pre-kindergarten through 12. - Intro: "compares ... with the official figures". Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Every bullet names its actor ("We added", "Microcosm now keeps").
- Retirement fix: "retirement income tax rules in 20 states" (some of
the 20 changed credits, not exclusions).
- Data bullets describe the pipeline change itself (district mapping,
usual hours, 2025 ASEC pool).
- Drop the state income-band bullet: microcosm#1102 adds the IRS
figures, but no calibration or release has used them yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Per Max, no wording mirrored from For All. A 4-word shingle comparison against TheAxiomFoundation/newsletters editions 2026-09-10, 2026-10-08 and 2026-10-29 now matches only proper names (BLS, Open Source Summit: AI Futures, Public Benefit Innovation Fund, the APPAM session title) and footer merge tags. Rewritten: the BLS lead sentence, the PolicyBench heading and paragraph, the Nava and Aspen sentences, and the Axiom section; its label and contents link now read "The Axiom Foundation". Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
policyengine.org/us/ballot-measures-2026 still returns 404 (the policyengine-app-v2#1415 proxy has not merged), so the seven dashboard links point at ballot-measures-2026.vercel.app, the deployment that proxy serves. All 31 URLs return 200. The PolicyBench audit sentence now says what the note says: flagged references checked against statutes and regulations. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
marked this pull request as ready for review
October 29, 2026 20:05
Contributor
Author
|
Merging as the record of what went out (sent 2026-10-29 20:05 UTC, Mailchimp 8a11e388cb, 762 non-UK recipients, on Max's go in chat). Head 5e91588. Gates: all 5 checks pass; MERGEABLE; GPT-6.1 Sol reviews r1 (dc6efb9), r2 (f0c2a82, ballot section) and r3 (24d3364, model updates) each returned SHIP AFTER FIXES, and the fixes are in 6e69c24, aec19c9 and 03bc70f. Later commits are Max-requested rewording (2461090) and the dashboard link repoint (5e91588; all 31 URLs 200 at send). Merge commit, per this repo's convention, so the pinned image commit 991a80d stays on main. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
US newsletter for Thursday, October 29, the day before the Prop 40 webinar.
Sent 2026-10-29 at 20:05 UTC (4:05pm ET) to 762 non-UK subscribers (Mailchimp
8a11e388cb, segment COUNTRY is not United Kingdom), on Max's go in chat: "send it to us (or non-uk), looksgood" (decision d2176). The file at5e91588is what went out. Archive: https://us5.campaign-archive.com/?u=e5ad35332666289a0f48013c5&id=8a11e388cbThe seven dashboard links point at
ballot-measures-2026.vercel.app/us/ballot-measures-2026/..., becausepolicyengine.org/us/ballot-measures-2026still returned 404 at send time (policyengine-app-v2 #1415, the proxy, is a draft). Both addresses serve the same deployment, so the vercel.app links keep working after #1415 merges.The last US newsletter sent was September 8 (Mailchimp
d19a2833b4). The September 30 draft (#26, Mailchimpb835a77fb1) never went out, so its lead (the graded poverty prediction) folds in here in short form. Once this edition sends, #26 can close as superseded.Sections
app/src/data/events.json), which dropped the no-position line and the days-before-the-vote line (app-v2 #1418, #1419). Register button: the Zoom registration link from the calendar event.assets/images/prop40-bridge-2026-10-29.png, pinned at991a80d)./us/california-wealth-tax/papers/web/), per Max: not posted to SSRN yet.slideshows/bls-taxsim-2026(README and config: Suitland, the three presenters, the topics). The 100+ attendance, about half in person, comes from the host's thank-you email of October 9./us/research/2025-spm-poverty-predictionand/us/research/2025-supplemental-poverty-measure(the senior miss falls from 0.76 to 0.09–0.26 points and the child miss from 0.92 to 0.6; the copy rounds these to 0.8, 0.1–0.3 and "about a third"), plus the September 15 recording and the Child Poverty Impact Dashboard.editions/2026-10-29.html)./us/research/policyengine-joss-paper(DOI 10.21105/joss.11115). The message house's do-not-claim on "peer-reviewed" exits once the paper has a DOI, and it now has one.After review
Max asked that nothing mirror the Axiom newsletter's wording.
2461090rewrote the BLS lead, the PolicyBench heading and paragraph, the Nava and Aspen sentences and the Axiom section. A 4-word comparison with For All's 2026-09-10, 2026-10-08 and 2026-10-29 editions now matches only proper names and footer merge tags.5e91588repointed the dashboard links and aligned the PolicyBench audit sentence with the note ("flagged reference answers against statutes and regulations").Template
US editions keep the light US layout from September 8 and the held September 30 draft (teal masthead rule, "View in browser", "Also in this issue", "Forward to a colleague"), not the dark UK template that CLAUDE.md describes. Images are pinned to commit SHAs. The template has no gradients, so the Outlook fallback note does not apply. The preheader text must match the campaign preview: "Tomorrow at noon PT, we put the Berkeley and Hoover estimates of Prop 40 in one open model and show where they part ways."
Review
An independent GPT-6.1 Sol review of
dc6efb9returned SHIP AFTER FIXES with nine findings.6e69c24applies them:text-transform: uppercase, matching the sent September 8 and May 19 US editions. Their source text is in sentence case.Rounds 2 and 3, same reviewer, on the sections added after Max's later asks:
f0c2a82), SHIP AFTER FIXES. It confirmed all six headline figures and the 20,694 coverage-gap figure againstmetrics.csv, and all seven dashboard paths against the dashboard's routes. Fixed inaec19c9:24d3364), SHIP AFTER FIXES. Fixed in03bc70f:Re-rendered at
03bc70fwith headless Chromium at 700px and 375px: no broken images and no horizontal scroll. At send (5e91588), all 31 unique URLs returned 200.🤖 Generated with Claude Code