HG · Metals / Demand-Side Macro Interaction
Z102090 – HG China Credit Impulse x Copper Seasonality
China credit expansion drives copper demand with a 3-to-6-month lag; the Q4-Q1 seasonal restocking window amplifies the effect. The interaction leads copper price direction when both conditions align.
At a glance
POST-SEASON SIGNATURE
Current reading
-0.777
504-day z of BIS China credit-gap 12m change x copper seasonality, tanh-bounded
Signal state
Moderate NEG
89 trading days in current state
Phase
STRESS
Positive equals credit and seasonal both strong, negative equals both weak
Sample
33 years
BIS credit history since 1993, roughly 8,266 trading days
As of Nov 04, 2026. Updated nightly. · NEXT WINDOW COPPER RESTOCKING WINDOW, NOVEMBER THROUGH MARCH
SIGNAL OVERVIEW
China drives copper demand on the long end of the horizon; nothing else in the macro set moves global refined-metal consumption the way Chinese credit does. But the impact is not uniform across the year. Physical-copper restocking in the Northern Hemisphere runs heavy from October through March; destocking weighs May through August. This signal takes the BIS China credit-impulse (the 12-month change in the nonfinancial-private credit-to-GDP gap, the canonical macro-finance measure) and multiplies it by a cosine-encoded day-of-year seasonality term. When strong China credit expansion aligns with seasonal restocking, the product reads near plus one. When weak China credit aligns with seasonal destocking, the product reads near minus one. In March 2009 the composite hit its maximum plus 1.0000 as the four-trillion-yuan post-GFC stimulus coincided with Q1 copper restocking; copper rallied aggressively through 2010-2011. Today the composite reads -0.777, in Moderate NEG territory.
LIVE CHART
24 years of Chile Atacama aquifer anomaly, read as water-stress regime for copper.
The dark line in the top panel is the composite, bounded between minus one and plus one. The pale red band at the top marks the regime where strong China credit coincides with seasonal restocking; this has historically preceded copper rallies at a 60-day horizon. The pale green band at the bottom marks weak credit × destocking, which has preceded muted or negative copper forward returns. The bronze line in the middle panel is the trailing 252-day rolling Spearman IC against copper forward 60-day returns. The lower panel is copper continuous back-adjusted price for context on the right axis in cents per pound.
COMPOSITE
+0.021 / tanh
STATE
Normal
PHASE
0
FWD 60d
+3.10% (baseline +2.32%)
SAMPLE N
8,764 days
TODAY’S PLAYBOOK READING
STATE
Moderate Neg
STATE AND PERCENTILE
Moderate NEG, 89 trading days running. The composite at -0.777 is in the band mapped by the engine.
WHY THE SIGNAL IS HERE
Regional aquifer storage over Chile’s copper-mining Atacama / Antofagasta region is at the level the engine has mapped to this state. The composite is a 63-day-smoothed, inverted-sign, 24-month z score, so the current reading reflects regional water balance against the trailing two-year baseline.
POSTURE FOR THE WEEKS AHEAD
Historical pattern is strongly as-is positive. State spread extreme-pos versus extreme-neg is plus 7.00 percentage points at 60d. The signal is strategic, not tactical.
WHAT THE CHART SHOWS
Today the composite reads -0.777, placing the dot inside the Moderate NEG band on the top panel. The BIS China credit data refreshes quarterly; intra-quarter moves in the composite come from the seasonality term as the day-of-year cycles. Watch the raw BIS release near the start of each calendar quarter for the atomic input.
TOP INPUTS FIRING
- BIS China credit-gap 12-month change Quarterly BIS release of nonfinancial-private credit-to-GDP gap, forward-filled to daily.
- Copper seasonality cosine Day-of-year encoded as cosine peaking around January 1 and troughing around July 1.
- Market phase direction Direction of the current state run, taken from the engine Market_Phase_Int column.
THE REGIME MAP
The three regime states, forward returns and sample share
Behavior across the three states computed from 8,764 engine trading days, with HG-conditional statistics against the 8,704-day subset where fwd 60d is observable. Baseline copper forward-60-day return (unconditional): plus 2.17 percent, median plus 1.56 percent, standard deviation 12.99 percent, percent negative 44.1 percent. State values from engine data, no fabrication.
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Credit and seasonal strong | +1 | 17.3% (1427 d) | +6.02% | +4.72pp | Jan 2023, Sep 2022, Feb 2021 | |
| Normal | 0 | 47.5% (3930 d) | +0.66% | -0.64pp | May 2026, Dec 2025, Mar 2025 | |
| Credit and seasonal weak TODAY | -1 | 35.2% (2909 d) | -0.18% | -1.48pp | Jul 2026, Mar 2026, Sep 2025 |
The historical pattern
24 years of Chile Atacama aquifer reading water-stress for copper
THE TRACK RECORD
Every state change in the last 2 years
Full state-transition history (several dozen transitions since 2002) is downloadable. The recent 2-year window shows the regime cadence into the current reading.
| State | Start | End | Duration | Fwd 60d COMEX HG |
|---|---|---|---|---|
| Extreme Pos Water Stress | 10/28/2024 | 02/20/2025 | 79d | +3.19% |
| Neutral | 02/21/2025 | 03/17/2025 | 17d | -0.85% |
| Extreme Neg Water Surplus | 03/18/2025 | 10/02/2025 | 138d | +7.62% |
| Neutral | 10/03/2025 | 12/31/2025 | 62d | +2.66% |
| Extreme Neg Water Surplus | 01/02/2026 | 02/27/2026 | 39d | -4.81% |
| Neutral | 03/02/2026 | 03/16/2026 | 11d | +9.34% |
| Extreme Neg Water Surplus | 03/17/2026 | 03/17/2026 | 1d | +9.98% |
| Neutral | 03/18/2026 | 04/23/2026 | 26d | +12.14% |
| Extreme Neg Water Surplus | 04/24/2026 | 09/16/2026 | 100d | +4.69% |
| Neutral | 09/17/2026 | 10/26/2026 | 28d |
Scroll inside the table for the full history. Forward 60-day COMEX HG is measured from each run’s start date; a reader cannot trade off a start-date they only identify in hindsight, so this is a transparency log, not an alpha claim. Asterisked (*) returns are partial. The run is less than 60 trading days old. The full 60-day comparable fills in as more data arrives.
What this measures
China consumes roughly 55 percent of global refined copper; Chinese credit cycles drive demand with a three-to-six-month lag. Copper also carries a seasonal pattern where physical restocking peaks November through March and destocking weighs May through August. The signal multiplies the BIS China credit-gap 12-month change by a cosine-encoded copper seasonality term, z-scores the product on a 504-day rolling window, and bounds through tanh.
Why it matters
China consumes about 55 percent of global refined copper. Chinese credit cycles lead copper demand by 3-to-6 months. But the effect is not uniform across the year; Northern-Hemisphere restocking concentrates physical demand into Q4-Q1. The interaction captures the strongest demand regimes when credit expansion meets restocking.
Strong readings above the plus 0.6 band have historically preceded copper fwd-60-day returns well above baseline, with a +6 percent mean. Pair with supply-side signals (water-stress composites, ACLED events) to confirm whether both sides of the market support the direction.
Copper supply shocks (Chilean drought, Peru political crisis, DRC permit shifts) can drive price higher even with a weak China credit × seasonality reading. The 2024-2026 window shows exactly this: composite negative, copper rose on supply tightness. The signal is a demand lens; pair with supply reads for a two-sided view.
GO DEEPER
What the Chile Atacama Aquifer composite actually tells you (and what it does not)
The brief covers the BIS credit-impulse construction and the cosine-encoded seasonality term, three signature episodes (2009 Q1 post-GFC stimulus series-max, 2015-2016 China hard-landing concerns, 2024 series-min), and four caveats on reading a quarterly-input macro signal as a daily tape. Includes regime-conditional forward returns and the 60-day horizon decomposition.
Signal consumers and siblings
Not investment advice · The analysis on this page is informational and educational. Signal readings, historical case figures, and empirical statistics are derived from publicly available data and are provided without warranty. Hypothetical performance results have inherent limitations, including the benefit of hindsight. Readers who intend to trade should consult a licensed investment professional and see the research brief for the full disclaimer.
Last refreshed Nov 05, 2026 08:35 PM ET