Methodology

Presidential Economic Pressure Index — Scoring Framework

PEPIndex · Methodology v1.2.13 · pepi.apiautomations.com

PEPIndex scores each policy event from 0 (full implementation) to 100 (total reversal or abandonment) across six weighted dimensions. The rolling composite is a 30-day exponentially-weighted moving average.

📄 Download the full methodology whitepaper (source-by-source lineage, scoring examples, calibration & validation, limitations, and version history) · validation & backtesting · changelog

1. What constitutes a PEPI event

A PEPI event is a public statement or formal action by the President that satisfies all four criteria:

Criterion	Definition
Specificity	Identifies a concrete action, target, or consequence — not vague rhetoric.
Actionability	Falls within presidential authority or influence to execute.
Measurability	Outcome can be independently verified: was the action taken or not?
Materiality	Carries economic, diplomatic, or policy consequences significant enough to warrant tracking.

2. Event domains

Domain	Scope
Trade & Tariffs	Import tariffs, trade deals, sanctions on goods, export controls
Foreign Policy	Military force threats, diplomatic ultimatums, territorial claims
Domestic Policy	Executive orders, government shutdowns, personnel and regulatory threats
Rhetorical	Lawsuits, boycotts, social media retaliation, interpersonal threats

3. Scoring dimensions

The composite score is a weighted sum of six dimensions:

Weight	Dimension	How it is measured
30%	Reversal Magnitude	How far the final action deviated from the original threat. For trade events: `(statutory − applied) / statutory × 100`. For non-trade: 7-point categorical scale.
20%	Timeline Slippage	Duration of deadline extensions, log-scaled. 7-day delay ≈ 40; 30-day delay ≈ 70; 90+ days ≈ 90+. No stated deadline = 30-point vagueness penalty.
15%	Rhetorical Dilution	NLP-measured shift in language intensity between initial threat and subsequent statements. Captures the linguistic softening that typically precedes a reversal.
15%	Scope Reduction	Percentage of originally targeted entities, countries, or product categories exempted from the final action. Directly informed by the statutory-vs-applied tariff rate gap.
10%	Framing Shift	Whether the outcome was reframed as a victory despite being a reversal — delta between administration characterization and independent analysis.
10%	Repetition Pattern	Whether this threat has been made and reversed before. Each prior reversal adds 20 points, capped at 100. Captures declining credibility.

Reversal Magnitude — trade events (live USITC data)

reversal_magnitude = (statutory_rate_announced − applied_rate_final) / statutory_rate_announced × 100

Worked example: 145% announced on China, 28.0% applied → 117pp gap → Reversal Magnitude = 80.7. Applied rates come from live USITC DataWeb ingest. The formula is bounded to [0, 100].

Composite formula

PEPI = 0.30 · Reversal + 0.20 · Slippage + 0.15 · Dilution + 0.15 · Scope + 0.10 · Framing + 0.10 · Repetition

4. Rolling index

The rolling PEPIndex is the 30-day exponentially-weighted moving average of all resolved event scores. More recent events are weighted more heavily with an exponential decay (~23-day half-life):

Index(t) = Σ [ score(i) · exp(−0.03 · days_ago(i)) ] / Σ [ exp(−0.03 · days_ago(i)) ]

Domain sub-indices (Trade, Foreign Policy, Domestic, Rhetorical) use the same formula filtered to events within each domain.

5. Source cadence

Source ingestion runs every 6 hours. The public composite index is published daily after validation; per-source freshness is exposed live at /api/v1/sources/status. Each component updates according to its official source release schedule — not all sources are daily.

Source	Role / used for	Official cadence
Federal Register API	Threat detection — Executive Orders, proclamations, USTR Section 301 notices	Daily (new documents)
CNN Truth Social Archive	Threat detection — primary presidential statement feed; repetition-pattern dimension	5-minute cache
USITC DataWeb	Outcome + economic measurement — applied rates (statutory-vs-applied gap)	Monthly
CourtListener	Outcome resolution — CIT and CAFC court rulings (primary automated ingest)	Every 6 hours
PACER PCL	Outcome resolution — CIT/CAFC case locator (automated fallback when CourtListener errors)	On CourtListener error
Congress.gov API	Outcome resolution — bill actions for legislation-linked events	Per bill-linked event
eCFR API	Outcome verification — point-in-time CFR snapshots for resolved events	Per resolved event
Gemini (Google Search grounding)	Outcome resolution — overdue non-trade events (independent reporting weighted above self-reported posts)	Per overdue event
BLS	Economic-impact measurement — four-group CPI proxies, PPI, employment	Monthly
FRED (St. Louis Fed)	Economic-impact measurement + macro context — CPI, trade balance, S&P 500, VIX, yields, dollar index	Daily / monthly (series-specific)
Yahoo Finance (yfinance)	Economic-impact measurement — SPY/VIX/sector-ETF reaction around threat/outcome dates	Every 6 hours

Threat classification uses Anthropic Claude; non-trade outcome reconciliation uses the grounded Gemini resolver. Both interpret public-source text into structured events and outcomes — they are not themselves data sources. See the full data-source taxonomy.

6. Economic impact measurement

For trade-domain events, PEPI measures real economic consequences using a four-group price-pass-through model. Live measurement runs every cycle on BLS CPI component series (apparel, household furnishings, food at home, medical commodities, services, core commodities) grouped into imported/domestic × affected/unaffected, with USITC applied rates and FRED pass-through/incidence series. The headline constants below are published, freely-citable figures; the retail-scanner replication package (≈359,104 products across five major U.S. retailers) was used only to verify methodology alignment, not as a live input.

Metric	Method	Value (2025 tariffs)
Retail pass-through rate	Applied-rate pass-through regression	~24% at 7 months
CPI contribution	Expenditure-weighted aggregation	+0.76pp all-items CPI
Consumer incidence	Border vs. retail incidence decomposition	~43% of tariff cost
Distributional gap	Within-category price quartiles	+1.5pp lower vs. upper quartile

7. Calibration events

Event	Domain	Authority	Score	Outcome
Section 232 Steel/Aluminum (2018)	Trade	Section 232	34.0	Partial reversal
Section 301 China / Phase One (2018–2020)	Trade	Section 301	53.0	Partial reversal
Canada/Mexico Steel USMCA Removal (2019)	Trade	Section 232	82.0	Full reversal
IEEPA 10% Baseline + Sec 122 (2025)	Trade	IEEPA, Section 122	91.0	Full reversal
Liberation Day: 145% on China (2025)	Trade	IEEPA	82.1	Partial reversal (28.3% applied)
14-Country Tariff Deadline (2025)	Trade	IEEPA	60.5	Delayed
North Korea: Fire and Fury (2017)	Foreign	—	95.5	Full reversal → summit
Afghanistan: Wipe Off the Earth (2019)	Foreign	—	97.0	Full reversal → peace deal
Greenland Annexation (2025)	Foreign	—	90.5	Full reversal → dropped
Russia: Extreme Consequences (2025)	Foreign	—	88.0	Reframed as summit
Pharma Tariffs 100% (2025)	Trade	IEEPA, Section 232	56.3	Delayed

8. Limitations

Rhetorical Dilution and Framing Shift depend on NLP models — classification errors are possible. Human review is required before events are confirmed.
Distributional impact (within-category price-quartile analysis) requires proprietary retail scanner data and cannot be replicated from public BLS aggregates alone.
USITC DataWeb applied rates cover HS-2 granularity for 11 countries. HS-10 resolution and additional country coverage depend on USITC data availability.
The rolling 30-day window means short-lived events have less persistent index effect than longer-running ones.
All sources are public, each with a defined role: threat detection (Federal Register, USTR notices, Truth Social), outcome resolution (USITC, courts, Congress.gov, eCFR, Gemini), and economic-impact measurement (BLS, FRED, market data). Static calibration constants are disclosed separately. No commercial aggregators, licensed market terminals, or third-party scrapers are used. See the data-source taxonomy.