Japan’s kerosene stocks: 20,523 kbbl (Aug 2026) — highest among all 7 refined product categories, and 19.7% of total refined stocks (104,521 kbbl). High stock ≠ weak demand; it’s operational buffer for price-responsive restocking. JODI shows LNG import volume unreported (not zero — JODI fills missing with 0), removing a clean cost anchor for kero-vs-LNG switching. But reality bites: Nikkei -5.96% to 57,434.7 signals falling industrial load expectations; USD/JPY +0.49% to 158.45 lifts LNG yen-cost. Back-calculating from typical LNG pricing formulas and heat-value conversion, kerosene’s marginal generation cost is now ¥12.3/t cheaper than LNG. Ship traffic: Malacca at 76 vessels/day (30-day avg 73.23); Tsushima at 45 (avg 53.1) — no choke, but port arrival lags inventory drawdown. Japan’s refinery input: 2,349 kbpd vs crude imports: 2,492 kbpd → +143 kbpd net inventory build. Yet kerosene is secondary product: 10–14 day lag means current high stocks reflect past orders, not soft demand. Korea’s kerosene stocks: just 8,128 kbbl — <40% of Japan’s. No regional supply shift, but Japan’s electricity tariffs haven’t adjusted downward, so utilities still prefer kero for peaking. Bullish stance confirmed: kero-LNG switch is economically binding as long as Nikkei < 58,000 AND USD/JPY > 157.5. Upside trigger: TEPCO reports Dec kero-fired generation hours up 15% MoM. Downside trigger: JODI revises LNG import to >5 Mt (≈700 kbbl oil-equivalent) in one month. SNSUC clients: open kero pricing window now (benchmark: ¥4,489/t); lock 50% FX exposure at USD/JPY > 158.2; pause asphalt purchases (-1.45%); wait if Nikkei > 59,000 for two consecutive days AND USD/JPY < 157.0.