04 · Sector Prosperity & Cycles
Half of industry analysis is "size and landscape" (slow variables); the other half is "where are we now" (fast variables). Cyclical industries earn from supply-demand mismatches; growth industries earn from rising penetration — both require judging prosperity position. This article dissects the nature of cyclical industries, the four signals of cycle position, classic buy/sell traits of cyclical stocks, the penetration rate curve, and when "prosperity investing" breaks down.
1. The Nature of Cyclicality: Supply-Demand Mismatch
Why cycles exist
The root of every cycle is that supply adjusts more slowly than demand: demand can change in a month, while supply (building plants, expanding capacity, commissioning lines) takes 1-3 years. Two mismatches therefore always recur:
| Mismatch | Result | Typical Manifestation |
|---|---|---|
| Demand > supply | Prices rise, profits explode | Windfall profits lure the whole industry into expansion |
| Supply > demand | Prices fall, profits collapse | Losses force tail capacity to exit |
💡 The cycle's self-perpetuating loop
The loop: windfall → expansion → glut → losses → clearing → supply contraction → price recovery → windfall. Studying a cyclical industry is essentially judging which link of this loop we are in.
Capacity cycles stacked on inventory cycles
| Cycle | Length (rule of thumb) | Driver | Watch Signals |
|---|---|---|---|
| Capacity cycle | 3-10 years | Capex, new capacity added or retired | Expansion announcements, construction-in-progress, utilization |
| Inventory cycle (Kitchin cycle) | ~3-4 years | Firms actively restocking or destocking | Absolute inventory levels, production-sales gaps, raw material prices |
- The capacity cycle sets the major direction (slow variable); the inventory cycle sets the swings (fast variable).
- Four inventory-cycle stages: active restocking (prosperity rising) → passive restocking (demand weakening, inventory piling up) → active destocking (prices falling) → passive destocking (demand recovering, inventory bottoming). Late active destocking plus early passive destocking is usually the cyclical bottom zone.
2. Judging Cycle Position: Four Signals
Signal 1: Product prices
| Condition | Meaning |
|---|---|
| Prices rising steadily + spot premiums/tightness | Prosperity upswing |
| Prices stalling at highs | Near the top — be alert |
| Prices below the whole-industry cost line | Bottom zone — wait for clearing |
| Prices flat below the cost line | Deep bottom, clearing underway |
⚠️ The most reliable bottom signal is "how long it stays," not "how far it fell"
The most reliable bottom signal is not how far prices fell but how long they stay below the industry-average cost line — only when the entire industry bleeds cash past its breaking point does capacity genuinely exit.
Signal 2: Inventory data
- Low upstream/midstream inventories + active downstream restocking = real demand, prosperity turning up.
- Bloated channel inventories + falling end-market prices = demand falsified; heavy destocking pressure.
- Inventory cycles lead price cycles: inventory peaks lead price peaks, and inventory bottoms lead price bottoms.
Signal 3: Capacity utilization
| Utilization Level | Verdict |
|---|---|
| > 90% | Supply tight; prices prone to rise |
| 75%-90% | Normal range |
| < 70% | Clear excess; price war highly likely |
| Industry-wide < 70% long-term | Clearing phase; wait for tail exits |
Signal 4: Capital expenditure signals
- Leaders expanding massively / waves of new entrants = prosperity peaking signal (supply about to be released).
- Capex going quiet / construction-in-progress completed then idled / industry M&A consolidation = supply contracting; the bottom nears.
- Expansion announcements always precede output by 2-3 years: today's expansion wave is the glut two to three years out.
Combining the four signals into a position call
| Combination | Position Verdict |
|---|---|
| Prices rising + low inventory + high utilization + expansion just starting | Mid-prosperity — the sweetest phase |
| Prices at highs + inventory accumulating + high utilization + expansion wave | Late prosperity — beware the top |
| Prices falling + high inventory + utilization sliding + expansions halted | Downturn — wait |
| Prices below cost line + destocking + low utilization + clearing happening | Bottom zone — positioning window |
3. Investing in Cyclicals: Buy at Losses, Sell at Windfalls
Bottom characteristics
| Characteristic | Description |
|---|---|
| Industry-wide losses | Most firms losing money quarterly; even leaders struggle near breakeven |
| Capacity clearing | Tail companies halt, go bankrupt, get acquired; capacity genuinely exits |
| Low inventories | Channel and producer inventories at historic lows |
| Prices near cost line | Commodity prices approach or dip below industry-average cost |
| Valuation signature | PE very high or undefined (losses), PB at historic lows |
| Sentiment signature | Analysts drop coverage; forums fill with curses; nobody discusses the sector |
Top characteristics
| Characteristic | Description |
|---|---|
| Expansion wave | Whole industry expands lavishly; cross-over entrants flood in ("pig farmers start mining") |
| Earnings peak | Quarterly profit hits record highs; collective market euphoria |
| Valuation signature | PE actually lowest (earnings peak maximizes the denominator), PB at historic highs |
| Sentiment signature | "This time is different" narratives emerge; everyone discusses the windfall |
| Warning signals | Product prices stalling, inventories accumulating, futures backwardation |
One-line core
Cyclical stocks: buy when the industry loses money, sell when it prints money; PE is a trap — PB and price position are the answer. Cyclical stocks often bottom at high PE (small denominator) and top at low PE (large denominator) — low-PE cyclicals are the classic value trap (see Case B in Article 03).
💀 Iron rule: low-PE cyclicals are the classic value trap
Buy cyclicals amid losses; sell them amid windfalls. They tend to bottom at high PE and top at low PE — so never value a cyclical by PE. Use PB and price position instead; when you see a "low PE" cyclical, run.
Operating discipline for cyclicals
- Bottom positioning means "tranches against sentiment": build positions in stages once clearing signals confirm, with an upper bound on averaging down.
- Top exits mean "watch price + watch expansion": peak profit ≠ sell point — exit when three signals resonate: stalling prices + expansion wave + inventory accumulation.
- Never go all-in on one sector: however right your cycle call, clearing timelines routinely miss by 1-3 years — position sizing is your only protection.
⚠️ Counterintuitive: peak profit is not the sell point
Peak earnings ≠ the sell point; exit when stalling prices, an expansion wave, and inventory buildup resonate together. The deadliest mistake in cyclicals is chasing record profits — that IS the top. The true sell point isn't when earnings look brightest, but when the market stops believing they can keep growing.
4. Prosperity in Growth Industries: The Penetration Rate Curve
Four penetration stages
| Stage | Penetration Range (rule of thumb) | Traits | Investment Focus |
|---|---|---|---|
| Introduction | 0-10% | Demand driven by early adopters and subsidies; volatile, unstable earnings | Theme/concept territory, high risk (see Article 05) |
| Takeoff | 10%-30% | Steepest penetration slope; volume and price rise together; high growth | Sweetest stage: earnings and valuation double gain |
| Maturity | 30%-70% | Growth decelerates, competition intensifies, price wars appear | Shift to landscape/leader logic (see Article 03) |
| Saturation | > 70% | Incremental growth capped; replacement-driven; growth turns cyclical | Valuation center drifts down; watch dividends and cash flow |
Why 10%-30% is the sweetest stage
- Enough certainty: crossing 10% proves the product has escaped the early-adopter "valley of death" — demand is validated;
- Plenty of headroom: multiples of growth remain before 70% saturation; slope steepest;
- Landscape unsettled but forming: heads emerging but not frozen — leaders' growth elasticity peaks here;
- Historical reference (illustrative): smartphones, EVs, renewable power generation — while penetration sat between 10%-30%, both industries and leaders dramatically outperformed the broad market.
Using the penetration curve — and its traps
- Spot the gear-shift point: beyond ~30%, growth typically drops from "explosive" to merely "high," and the valuation center falls with it — don't pay takeoff-phase valuations for maturity-phase growth.
- Penetration pitfalls: penetration = sales/potential users — change the denominator's definition and the conclusion flips; penetration inflated by subsidy-driven discounting (e.g., volume via rock-bottom pricing) can retreat once subsidies fade.
- Read volume and price separately: penetration up but unit price down (price war) doesn't guarantee revenue growth — penetration × unit price × price stability: all three matter.
5. Prosperity Tracking Checklist
Common industry data sources (illustrative — defer to official definitions)
| Industry | Core Data Sources | Key Indicators |
|---|---|---|
| Autos | Auto and passenger-car associations | Monthly sales, NEV penetration rate, dealer inventory |
| Real estate chain | Statistics bureau, property research firms | Sales area, new starts, completions, secondhand transactions |
| Steel | Steel associations and data portals | Blast-furnace operating rates, inventory, per-ton profit |
| Nonferrous metals | Metals exchanges and associations | Spot prices, inventory, processing fees |
| Chemicals | Chemical information services | Product spreads, operating rates, inventory |
| Coal | Industry associations, port data | Port inventory, daily consumption, contract prices |
| Semiconductors | Industry bodies, SIA | Sales YoY, inventory, equipment billings |
| Consumer | Retail-sales statistics, baijiu wholesale/e-commerce data | Retail growth, sell-through, inventory cycles |
Universal tracking framework (monthly refresh)
| Category | Indicators | Signal Meaning |
|---|---|---|
| Price | Product prices, futures spreads, spot premium/discount | Direct reflection of marginal supply-demand |
| Inventory | Producer, channel, downstream inventories | Leads price cycle by 1-3 months |
| Capacity | Utilization, new capacity commissioned, exit announcements | Supply trend |
| Demand | Sales volumes, orders, exports | Reality check on demand |
| Policy | Subsidies, standards, export controls | Exogenous shocks to demand/supply |
💡 Data tracking lives on continuity
Continuity is everything: one month's swing is noise; only the 3-6 month slope is trend. Turn indicators into trend calls, not point calls.
6. When "Prosperity Investing" Breaks Down
Why peak prosperity is precisely when danger peaks
Prosperity investing (chasing high-momentum industries) works well through an industry's early and middle phases, but systematically fails near the prosperity peak:
| Failure Mode | Mechanism |
|---|---|
| Crowded-trade unwind at the prosperity peak | When earnings look best and logic flows smoothest, the whole market piles in; the moment growth decelerates at the margin (even if still high), valuation de-rating strikes first, earnings cuts follow |
| "Growth slowdown = stock halves" | Prices react to marginal change, not absolute level — growth easing from 60% to 30% still triggers de-rating |
| Prosperity falsified | High momentum built on subsidies, channel stuffing, or one-off demand collapses the instant the data wobbles |
| Top-calling fails | Tops can't be timed precisely; after the expansion wave starts there may be another year of gains — but those are profits on a knife's edge |
Three principles to avoid failure modes
- Distinguish where you are within prosperity: participate in early-to-mid upswings; once expansion wave + inventory buildup + stalling prices appear, downgrade to observation no matter how festive things look.
- Distinguish prosperity from landscape: industries combining rising prosperity with deteriorating structure offer weak, short-lived profit elasticity (see Case B in Article 03).
- Keep a margin of safety: hot industries' valuations usually pre-spend 1-2 years of earnings — anchor your purchase on next year's valuation, not this year's results.
7. Reading Cycles and Growth Together
Many real industries mix the two — "growth inside cycles, cycles inside growth" (EVs, display panels, semiconductors):
| Industry Type | Primary Conflict | Primary Toolkit |
|---|---|---|
| Strongly cyclical (steel, coal, shipping) | Supply-demand mismatch | Price, inventory, utilization, capex |
| Growth + cyclical (semiconductors, panels, renewables) | Penetration gains × capacity cycle | Penetration curve overlaid with four cycle signals |
| Pure growth (software, some consumer) | Penetration and landscape | Penetration rate, average ticket, share |
💡 The "two-layer judgment" for hybrid industries
Hybrid industries need two layers: first judge the penetration-rate direction (growth core-holding logic), then judge the short-term supply-demand sub-cycle (cyclical swing logic). Stack size when both layers point the same way; use small positions to harvest cycle swings when they diverge.
⚠️ Risk Warning
⚠️ Risk Warning
Cycle and prosperity judgment carries the largest forecasting errors in all of industry research — bottoms can last years ("there's a basement beneath the floor"), tops can run crazier and longer than you imagine; inventory and utilization data lag and carry definitional padding; penetration estimates are estimates, not facts. More importantly, prosperity investing loses the most money exactly at emotional peaks: chasing hot momentum + crowded institutional trades + pre-spent valuations is the recipe behind most historic "track halvings." This is educational methodology content, not investment advice; hedge every prosperity call with position sizing and stop losses.