Economic Indicators

Economic Indicators for Trading

When to Activate

• Interpreting economic data releases and their market implications
• Building nowcasting models or tracking economic momentum
• Trading around data releases (NFP, CPI, PMI, GDP)
• Understanding leading, coincident, and lagging indicator frameworks
• Analyzing surprise indices and market reaction patterns

Core Concepts

Indicator Classification

Leading Indicators (predict future economic activity):

• PMI (Purchasing Managers' Index) — survey-based, 50 = expansion/contraction threshold
• ISM Manufacturing PMI: released first business day of month
• ISM Services PMI: released third business day of month
• Markit/S&P Global PMI: flash estimate mid-month (earlier signal)
• Sub-components matter: new orders (most forward-looking), employment, prices paid (inflation signal)
• PMI below 50 for 3+ months historically precedes recession
• Yield curve slope — 3m10Y spread, most reliable recession predictor
• Building permits — residential construction leads broader economy by 6-12 months
• Initial jobless claims — weekly, high-frequency labor market signal; sustained rise above 300K signals trouble
• Consumer confidence — University of Michigan, Conference Board; forward-looking spending intentions
• Stock market — S&P 500 is officially a leading indicator (part of LEI composite)
• Money supply (M2) — growth rate leads nominal GDP by 6-18 months

Coincident Indicators (confirm current economic state):

• Nonfarm Payrolls (NFP) — gold standard for labor market, released first Friday of month
• Industrial production — factory output, mining, utilities
• Real personal income — inflation-adjusted income excluding transfers
• Retail sales — consumer spending proxy (70% of US GDP)

Lagging Indicators (confirm trends after the fact):

• Unemployment rate — peaks after recession ends (lagging by 3-9 months)
• Core CPI — inflation responds to economic conditions with a lag
• Corporate profits — reported quarterly, backward-looking
• Average duration of unemployment — rises after recession, slow to recover
• Bank lending — credit standards tighten after downturn begins

Key Indicator Deep Dives

Nonfarm Payrolls (NFP):

Release: First Friday of month, 8:30 AM ET
Components that matter:
1. Headline number (consensus vs actual)
2. Prior month revision (often +/- 50K, markets react to direction)
3. Unemployment rate (U-3) and underemployment (U-6)
4. Average hourly earnings (YoY) — wage inflation proxy
5. Average weekly hours — leading indicator within the report
6. Labor force participation rate — structural employment picture

Market reaction framework:
- Strong NFP + high wages: hawkish (rates up, USD up, stocks mixed)
- Strong NFP + moderate wages: Goldilocks (stocks up, USD modest)
- Weak NFP + falling wages: dovish (rates down, USD down, stocks mixed)
- Weak NFP + rising wages: stagflation concern (stocks down, rates volatile)

Typical market impact:
- S&P 500: 0.5-1.5% move on significant surprises
- 2Y Treasury: 5-15 bps move
- EUR/USD: 50-100 pips move

CPI (Consumer Price Index):

Release: ~13th of month, 8:30 AM ET
Key measures:
1. Headline CPI (includes food and energy — volatile)
2. Core CPI (excludes food and energy — Fed focus)
3. Supercore (core services ex-housing — Powell's preferred measure)
4. Shelter/OER — largest component (~35%), lags real-time rents by 12 months
5. Trimmed mean CPI (Dallas Fed) — removes outliers for underlying trend

Monthly vs annualized:
- 0.2% MoM core = ~2.4% annualized (near target)
- 0.3% MoM core = ~3.7% annualized (above target)
- 0.4%+ MoM core = ~4.9%+ annualized (significantly hot)

Market sensitivity: CPI has become the most market-moving release
since 2022 inflation surge. Single 0.1% surprise in core can move
S&P 500 by 1-2% and 2Y yields by 10-20 bps.

PMI (Purchasing Managers' Index):

Scale: 0-100, 50 = neutral
Above 50: expansion
Below 50: contraction
Above 55: strong expansion
Below 45: significant contraction (recession territory)

Sub-indices (ISM Manufacturing):
- New Orders: most forward-looking, leads headline by 1-2 months
- Production: current activity level
- Employment: labor market signal
- Prices Paid: input cost inflation (leads PPI/CPI)
- Supplier Deliveries: supply chain stress (higher = longer delays)
- Inventories: stock building vs destocking cycle
- New Export Orders: global demand signal

Global PMI composite provides real-time GDP proxy:
PMI 50 ≈ 0% GDP growth
PMI 55 ≈ 3% GDP growth
PMI 45 ≈ -2% GDP growth (recession)

Surprise Indices

Citi Economic Surprise Index (CESI):

Measures: weighted average of data surprises (actual - consensus)
Positive: data is beating expectations (economy stronger than expected)
Negative: data is missing expectations

Trading signal:
- CESI rising from negative to positive: risk-on, pro-cyclical assets
- CESI falling from positive to negative: risk-off, defensive assets
- CESI is mean-reverting: extreme readings tend to reverse within 2-3 months
- Not a level indicator — measures second derivative of expectations

Limitations:
- Consensus estimates adjust over time (expectations catch up to reality)
- Mean reversion is mechanical, not predictive
- Works better for FX (rate expectations) than equities

Bloomberg Economic Surprise Index:

• Similar concept, different weighting methodology
• Available for US, Europe, China, EM

Nowcasting Models

GDP Nowcasting:

Atlanta Fed GDPNow:
- Real-time GDP estimate updated as data releases occur
- Mechanical model: no subjective adjustments
- Starts inaccurate early in quarter, improves as more data arrives
- Useful for tracking direction, not precise level

NY Fed Nowcast:
- Uses larger dataset, dynamic factor model
- Updated weekly
- Generally more stable than GDPNow

Market application:
- Compare nowcast to consensus GDP forecast
- If nowcast >> consensus: growth surprise likely, bullish positioning
- If nowcast  USD supportive
- Europe weakening -> ECB likely to cut before Fed
- China recovering -> commodity demand positive
- Global growth: mixed, not synchronized

Trade ideas:
1. Short EUR/USD: Europe/US divergence, rate differential widening
2. Long copper: China demand recovery + constrained supply
3. Overweight US vs European equities: growth differential

Monitor: if US PMI breaks below 50, divergence thesis changes

Example 3: Nowcast vs Consensus Gap

Situation:
- Q3 GDP consensus: +2.0% annualized
- Atlanta Fed GDPNow (8 weeks into quarter): +4.5% annualized
- NY Fed Nowcast: +3.8% annualized
- Gap: nowcasts are ~2.0% above consensus

Interpretation:
- High-frequency data is significantly stronger than economists expect
- Consensus will likely be revised higher as more forecasters update
- GDP release may surprise significantly to the upside

Trading implications:
- Growth surprise = hawkish for rates (less likely Fed cuts)
- Positive for cyclical sectors (financials, industrials)
- Negative for long-duration bonds
- USD supportive (higher rates, growth differential)

Position: overweight cyclicals, underweight duration
Monitor: watch for retail sales and industrial production to confirm

Quality Gate

Before trading around economic indicators, verify:

• [ ] Release date and time are confirmed (holidays and government shutdowns can delay releases)
• [ ] Consensus estimate is from a reliable source (Bloomberg, Reuters) with sufficient forecaster coverage
• [ ] Prior revisions are noted — direction of revisions often matters as much as the headline
• [ ] Positioning data is checked — CFTC, options positioning, rate expectations via Fed funds futures
• [ ] Seasonal adjustment methodology is understood — non-seasonally adjusted data can mislead
• [ ] Sub-component analysis is prepared — headline can mask important shifts in composition
• [ ] Market reaction function is defined before the release — what will you do in each scenario?
• [ ] Risk management is set — stops, position limits, and max drawdown defined for event trades
• [ ] Cross-market correlations are mapped — rates, FX, equities, and commodities all react together
• [ ] Historical reaction patterns are reviewed — how has the market reacted to similar surprises in the past?

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: brainbytes-dev
• Source: brainbytes-dev/everything-claude-trading
• License: MIT

Install and usage instructions live in the source repository linked above.