PolicyEngine / PolicyEngine/policyengine-scorecard

Tariff tracker claim family: harvest, registry, and the PE-machinery computability gate

Open
#17 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
0
Forks
0
Avg merge
6d 12h
Merged PRs (30d)
32

Description

Tariff tracker claim family: harvest + registry + computability gate

The Axiom US tariff encoding reached conformance against the Yale Budget Lab tracker's statutory panel (axiom-oracles us-tariff-yale, 2026-08-03), which opens the estimation-layer question this repo owns: the tariff trackers publish weighted aggregate claims that belong here as external claims, not in axiom-oracles (whose legs are statutory-only — routing settled with Max 2026-08-03).

Claim family to harvest/verify in staging (the 8/2 US harvest covered these sources for tax-model claims; confirm whether their tariff-tracker surfaces were captured, and add them if not):

  • Budget Lab "State of U.S. Tariffs" (Jul 24 vintage: avg effective rate 11.1%; ~$1,100/household/yr; 0.7% consumer prices; ~$1.9T/10yr revenue; distributional tables) + their daily tracker series (by country/authority/GTAP, sha256-manifested releases).
  • TPC "Tracking Trump Tariffs" (9.3% average — different basis than BL's 11.1%; country/product weighted rates; $581B/2026-36 score of the 2026-07-23 forced-labor §301 finalization).
  • PWBM effective tariff rates and revenues series; Tax Foundation tracker; Bipartisan Policy Center revenue tracker.

Registry per doctrine (#6): validation_comparator facts in Ledger; calibration_relationship = held_out for all of these (none are consumed as targets); per-row basis metadata is load-bearing (pre/post-substitution, exposed vs all-pairs, statutory vs collected — the trackers themselves differ ~2pp on headline ETR by basis alone).

Computability gate: these rows become computable comparisons only once the PE tariff machinery exists (Axiom statutory wedge artifact → Census import weights → consumption-category price wedges → household incidence; the planned PE tariff project). Until then they stage as harvested claims. Descriptive-by-default framing throughout per #9 — divergence vs the trackers is methodological_difference unless tied to a known issue (the two reference-side issues filed upstream — Budget-Lab-Yale/tariff-rate-tracker#27 §201 staged-table end, #28 column-2 registry — are citable action_links for their respective divergence classes).

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the claim-family requirements, the Ledger registry doctrine in #6, and the computability gate described for the planned PE tariff project. Check the listed Budget Lab, TPC, PWBM, Tax Foundation, and Bipartisan Policy Center tracker surfaces and the cited upstream issues. Done means tariff-tracker claims are harvested with basis metadata, staged until the PE machinery exists, and classified descriptively with appropriate action links.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data-engineering
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.