Playbook › Playbook
A disclosed ASP series settles cyclical-vs-structural when revenue alone cannot
Claim
Revenue is price × volume, and a revenue decline is ambiguous between the two. For a commodity component supplier that ambiguity is the investment question: falling price is share loss and margin destruction that does not recover; falling volume after a shortage is inventory arithmetic that does.
Diodes Incorporated discloses both changes, every year, in the MD&A. Chain-multiplying them produces an index that decides the question outright.
| Year | Disclosed ASP change | Volume change | Units (B) | Revenue ($M) | ASP index (2021=100) |
|---|---|---|---|---|---|
| 2021 | — | — | 58 | 1,805 | 100.0 |
| 2022 | +28.5% | −13.8% | 50 | 2,001 | 128.5 |
| 2023 | −2.7% | −16.0% | 42 | 1,662 | 125.0 |
| 2024 | −14.9% | −7.3% | 39 | 1,311 | 106.4 |
| 2025 | −1.7% | +15.0% | 45 | 1,482 | 104.6 |
Revenue fell 26% from the 2022 peak and gross margin fell 10 points. The obvious reading is commodity price destruction by Chinese competitors. The ASP index says 2025 prices are 4.6% ABOVE 2021 — every dollar of the decline is the unwind of a +28.5% shortage spike, and price never returned to the pre-shortage baseline, let alone fell below it.
Validation: chain-multiplying the disclosed ASP and volume changes reproduces reported revenue to within 1% in every year. The series is internally consistent, which is what makes it usable rather than merely suggestive.
Why it works
The shortage inflates the base year, and every subsequent comparison is against that inflated base. A revenue or margin chart anchored on the peak therefore cannot distinguish "we gave back a windfall" from "we are being undercut." The ASP index re-anchors on the pre-shortage year, which is the only base that represents normal competitive conditions.
The same decomposition also exposes what is structural. DIOD's units went 58B (2021) → 45B (2025), −22% while price held. That volume loss — in the low-ASP consumer/computing tail — is real and is not coming back. Price and volume degrade for different reasons and recover on different schedules; a series that separates them lets you say which happened.
How to apply
- Before concluding "commodity price erosion," search MD&A for a disclosed ASP or average-selling-price change. It is more common than expected among discrete/analog, connector, passive and packaging suppliers, and it is almost never picked up by sell-side or vendor screens.
- Chain-multiply to a pre-shortage base year (2019–2021 for the semi cycle), not to the peak. Anchoring on 2022 reproduces the error you are trying to avoid.
- Validate the series by checking that ASP × volume reproduces reported revenue. If it does not reconcile, the disclosures cover different scopes and the index is unusable.
- Read management's own attribution of the ASP move. DIOD attributed its worst year (−14.9%, 2024) to mix — the high-ASP automotive and industrial segments shrank as a share, dragging the blended average — not to competitive pricing. A mix-driven ASP decline reverses when mix reverses; a competition-driven one does not.
- Pair it with the peer test. If every peer declined, no one gained the share the bear case requires. Compare gross-margin damage, not revenue depth: a company surrendering price to hold sockets bleeds margin faster than peers. DIOD's revenue fall was the deepest of nine peers but its margin damage was mid-pack — which is the signature of volume give-back, not price surrender.
- The finding does not make the stock cheap. DIOD's ASP series established that the decline was cyclical; the stock was still ~39% above fair value because the recovery had already happened. Separating price from volume tells you what the earnings base is, not what to pay for it → [[pitfall-multiple-trim-inverts-on-peak-cycle-cyclicals]], [[principle-down-a-lot-is-not-cheap]].
Related: [[pattern-shortage-pricing-masks-share-loss]] (the inverse case — revenue growth on falling unit share) · [[principle-primary-source-beats-vendor]] · [[pitfall-divested-segment-corrupts-multiyear-cagr]]
History
- 2026-08-06 — Found during
/analyze DIODwhile trying to size the Chinese-competition bear case. The China thesis had strong priors and weak evidence; the ASP series was what actually settled it.static— the technique is a property of the disclosure, not of any cycle.