Stood up from the investigation 2026-06-18-china-structural-discount-kweb-fxi-multiple-not-earnings (user-flagged 2026-05-06; Theo's 2026-04-16 china-tech-rebound orphan folded in + closed). Watchlist china-tech.json created (KWEB/FXI/ASHR/BABA/PDD/BIDU/JD/NTES); ACTIVE.json + this log.

Core reframe: it's a multiple problem, not an earnings problem — ~15% consensus 2026 MSCI-China EPS growth while KWEB sits ~flat over a decade and −26% over 6mo into oversold. The discount is the sum of three stackable structural caps (VIE/governance + CCP-arbitrary-action tail + macro overhang); an earnings beat fixes none of them.

Data at birth (2026-06-18): complex washed out and in a downtrend — KWEB RSI 39 / −26% 6mo, BABA RSI 21 / −26%, PDD −24%, BIDU RSI 28; all below SMA20. Key isolation tell: ASHR (onshore, no VIE) is ~−3% off its 52wk high while KWEB (offshore VIE) is ~−26% — the VIE/governance discount is real and specific, not generic China weakness.

Posture: monitoring, not a buy. Per trend-hold, oversold-in-a-downtrend = falling knife; a re-rate needs a multiple-ceiling lift (VIE de-risk / sustained multiple expansion / macro turn) confirmed by KWEB reclaiming a rising SMA on volume — not an oversold bounce. Real book holds none.

Primary gauge going forward: the KWEB↔ASHR spread (narrowing = VIE discount easing).

🔄 2026-06-25 full-scan tape check: CONFIRMING (discount #1 isolated) — ASHR (onshore, no VIE) strong-up RSI 58 (−1.7% off high) while the offshore-VIE wrapper collapses (KWEB RSI 24, FXI 23, BABA 17, −50% off high) — the KWEB↔ASHR spread WIDENING is the further-discount trigger and confirms the "VIE-wrapper is what's discounted" call exactly. Not an entry (oversold ≠ setup, no volume turn). (Data tables spliced fresh to the 2026-06-25 close; full cross-scan read in 2026-06-25-full-scan-market-brief §6.)