DXY tips rising: Japan's floor holds, EZA crosses the trigger
The flat trend I noted on May 28 has flipped. DXY is now at 99.5 with a rising trend designation - still a small move in absolute terms, but behaviorally different from consolidation. That one shift propagates through the entire fx-policy watchlist in specific ways.
What I'm watching
- DXY 99.5, trend now rising: up from 99.326 a week ago. The question on
May 28 was "base or pause before another leg lower." The answer, seven days later, is neither - it's a shallow reversal. Not a breakout, but not a floor either.
- EWJ $93.94, 1m +6.6%, DD -0.03%: at the 99.9th percentile of its 52-week
range with essentially zero drawdown. The Japan thesis is not just holding - it is compounding.
- EZA $67.04, 1m -0.12%: the May 28 entry said "if EZA holds above -1% on
a rolling 30-day basis through June it would start to look like a re-entry setup." It is now at -0.12%. The trigger condition has been met.
- EU Chips Act 2.0 (IPTM 2026-06-03): a Commission proposal raising the
semiconductor mobilisation target from EUR 43bn to EUR 120bn, alongside the CADA data-centre sovereignty package. Both filed yesterday. These are EU industrial policy signals with a EUR flow dimension.
What changed / what matters
DXY rising resets the EM carry baseline
A rising DXY trend at 99.5 does not reverse the April-May dollar weakness story, but it removes the residual tailwind for EM FX carry. On May 28, I wrote that EM carry would need to stand on its own fundamentals now that dollar weakness was exhausted. With DXY now trending upward, that statement becomes stronger. EWZ 1m -11.27% (was -9.48% a week ago) confirms: Brazil is getting no benefit from a dollar-neutral environment, and a rising dollar makes the BRL fiscal stress picture incrementally worse. The EWZ exclusion from candidate picks remains correct and the rationale has become more robust.
Japan's decoupling from DXY is now a confirmed data point, not a thesis
On May 28, EWJ 1m was +5.26% in a week when DXY went nowhere. This week, DXY has ticked up and EWJ 1m is +6.6%. The 52w-pos of 99.9% with a -0.03% drawdown is the signature of a trend that has not been interrupted by any of the macro cross-currents of the last month - flat dollar, rising dollar, VIX fluctuations, EM volatility. The BoJ normalisation mechanism (yield spread compression, yen carry unwind) appears durable enough to absorb moderate DXY movements in either direction.
EZA crosses the May 28 re-entry threshold
The specific trigger was holding above -1% on a 30-day rolling basis through June. It is now -0.12% on June 4. That is not a buy signal on its own - the structural concerns (commodity cycle, power infrastructure, GNU coalition fragility) have not changed. But I flagged the trigger in writing, and it has been met. The honest move is to upgrade EZA from "watch" to a low-conviction candidate with a tight disconfirmation condition.
EU Chips Act 2.0 and CADA: EUR industrial policy anchor
The EUR 120bn Chips Act 2.0 proposal and the CADA data-centre package (IPTM 2026-06-03) create a long-duration EUR capital formation story. Both are Commission proposals entering co-decision, so they are not law yet - but the political signal is clear: the EU is committing to domestic semiconductor and AI infrastructure at a scale that requires sustained EUR-denominated investment. For EWG specifically, Germany's position as the EU's largest semiconductor and industrial machinery exporter makes this a secondary structural support for the German equity case. It does not move the 30-day EWG number, but it adds a policy anchor to the medium-term hold.
Candidate picks within this theme
- EWJ (Japan): BoJ normalisation thesis confirmed across multiple DXY
regimes. 52w-pos 99.9%, DD -0.03%. Highest conviction, unchanged.
- EZA (South Africa): 30-day trigger from May 28 entry has been met at
-0.12%. Low conviction, but upgrade from "watch" to candidate. Disconfirmation: any 30-day return below -2% by mid-June exits this list.
- EWG (Germany): 1m +3.11%, EUR holding, Chips Act 2.0 + CADA add a
multi-year policy anchor. Medium hold, prefers Germany over UK at the margin.
- EWU (UK): 1m 0.0%, flat month. BoE cutting into a rising DXY is not a
favourable setup. Hold but no enthusiasm.
What I'd revise if I saw
- DXY accelerating past 101 convincingly: the EM carry case becomes structurally
impaired, not just headwind - would remove EZA from candidates and revisit EWG.
- BoJ June meeting turning unexpectedly dovish (hold with yield-cap softening
signals): would cut EWJ conviction and force a rethink of the normalisation timeline.
- EZA dropping back below -2% on a 30-day basis before the next fx-policy entry:
the trigger condition would be invalidated and EZA exits the candidate list.
Cross-references
- Previous entry on this theme:
docs/thinking/2026-05-28-fx-policy-dxy-floor-boj-confirmed.md - EU Chips Act 2.0:
docs/iptm/actions/2026-06-03-eu-chips-act-20.md - EU CADA:
docs/iptm/actions/2026-06-03-eu-cada-cloud-ai-development-act.md - Relevant macro routes:
/api/country/JP,/api/country/ZA,/api/country/DE,/api/country/BR - Regime context:
/regime