04 · Industry Data: Supply-Demand Analysis and Inventory-Cycle Reading

Macro data tells you "how the economy is doing"; industry data tells you "how your instrument is doing". Crude oil, copper, hogs, real estate, new energy — every industry has a "data map": inventories, output, prices, orders. This article teaches you to translate that data into supply-demand judgments and price signals, using two core methods: the inventory cycle's four stages and volume-price divergence.

1. The Data Systems of Cyclical Industries

Crude Oil

Copper

Hogs

Real Estate

New Energy (EV Batteries / Solar as Examples)

2. Reading the Inventory Cycle

The inventory cycle (Kitchin cycle) is the core framework linking supply-demand data to prices. Prices are not determined by inventories, but the inventory cycle amplifies supply-demand forces.

Judgment points:

• The stages are classified by the combination of "demand direction × inventory change", not inventory levels;
• Key leading sequence: new orders (demand) move first → inventories next → price last;
• Buying early in active restocking and late in passive destocking are the two classic cyclical entry points; late passive restocking and active destocking are reduction phases.

3. Volume-Price Divergence: The Industrial Meaning

Operational implications:

• "Low inventory + rising price" is the most fundamentally solid long window;
• "High inventory + falling price" is the most solid short/avoid window;
• When inventory and price move in the same direction (both up or both down), let inventory lead and price confirm — inventory is the precursor, price the confirmation.

4. Timeliness and Lead-Lag of Industry Data

Why "read the data before the price": price lags supply-demand; data expresses it concurrently or ahead of time.

5. What "High-Frequency Data" Means

Official statistics (monthly/quarterly) are too slow; institutions use weekly/daily high-frequency data to sense the economy early:

• Official data (monthly/quarterly) lags publication; high-frequency data flags turning points 2–6 weeks before official releases;
• The "grassroots surveys" and "industry-chain research" in sell-side reports are essentially high-frequency data;
• For individuals: official data + industry association data + exchange inventories + commodity prices already cover 80% of high-frequency information needs.

6. An Industry-Data Interpretation Workflow

① Pick your industry    → go deep on only 1-2 industries; don't overreach
② List core indicators  → pick 5-8 per industry (see Section 1)
③ Track on fixed days   → 30-60 minutes weekly, updating on calendar rhythm
                          (one weekly table, one monthly table)
④ Write observation notes → monthly note: data changes, inventory-cycle stage call,
                          volume-price conclusions
⑤ Cross-check vs. price → observations vs. futures/equity trends; gaps = opportunity or risk

Weekly tracking template:

7. Case Study: A Full Walkthrough of the Hog Cycle

Applying the earlier framework end-to-end (teaching-convention figures, method demonstration only):

Background data (fictional starting point):

• Breeding-sow inventory: 42 million head (normal capacity band roughly 39–44 million);
• Average hog price: CNY 14/kg (industry breakeven ~15.5/kg, deep losses for farmers).

Step 1: judge the direction of supply — breeding-sow inventory leads slaughter supply by 10–14 months:

Step 2: locate the cycle stage — combine with the "demand × inventory" frame:

Step 3: cross-check against price and defuse risks:

• If hog prices have risen but the sow herd keeps falling → insufficient restocking, the rally is more sustainable;
• If prices just turned and the sow herd rebounds quickly → capacity rebuilt too fast, the downturn may arrive early;
• Policy variable: central/local pork stockpiling programs typically step in at price troughs — such announcements themselves signal a "policy bottom", but scale is limited: they support, not reverse, the trend.

8. Common Mistakes in Interpreting Industry Data

Risk Warning