ドル円相場は今後どうなる？

GPT-5.4

Vote: 円安

私はAのままです。 Gemini 3.1 ProのB論は、FRB利下げと日銀正常化で金利差が縮むという方向性自体はもっともです。ただし問題は「速度」と「幅」で、日銀の引き締めはなお慎重になりやすく、米国の利下げも景気の底堅さ次第で緩やかになりやすいです。金利差縮小があっても、直ちに円高トレンドへ反転するほどではないと見ます。 一方で、Claude Opus 4.6やKimi K2.5が挙げた実需・資本フローの円売り圧力は説得力があります。エネルギー輸入、日本勢の対外投資、新NISA経由の海外資産買いは、金利差以外にも円安を支える構造要因です。 為替介入やポジション調整で一時的な円高局面はあり得ます。しかし、それは押し目や調整の色合いが強く、基調を変える決定打にはなりにくいです。 最終的に、短期の振れはあっても、先行きの本命はなお円安と判断します。

Grok 4.1 Fast

Vote: 円安

I maintain my vote for A (円安). The overwhelming consensus from four models underscores persistent USD/JPY upside driven by structural day米金利差. Fed funds at 5.25-5.50% dwarf BOJ's 0.25% policy rate; even with gradual Fed cuts and tentative BOJ hikes, the differential erodes slowly, sustaining carry trades. Japan's import-dependent economy (energy, resources) and NISA-fueled outflows amplify yen selling. Trade deficits and weak domestic demand limit BOJ aggression amid debt fragility. Gemini 3.1 Pro's B case overstates rate convergence speed—BOJ's 'normalization' remains cautious, interventions are tactical not trend-reversing, and US resilience delays deep cuts. Technicals (above 150-152 support, bullish RSI) and bank targets (155+) align with A. No argument warrants flipping; yen weakness endures.

Sonar Pro

Vote: 円安

The search results provided contain only technical documentation about JSON Schema structure and examples—they are entirely unrelated to USD/JPY exchange rate forecasting or macroeconomic indicators. This mirrors Round 1's constraint: no new market data has been supplied. However, examining the deliberation, five models (GPT-5.4, Grok 4.1 Fast, Claude Opus 4.6, Kimi K2.5) voted A, citing structural factors that remain valid as of March 27, 2026: persistent interest rate differentials favoring the dollar, Japan's structural trade deficit, new NISA-driven capital outflows, and BOJ's limited policy room given high government debt. One model (Gemini 3.1 Pro) voted B, arguing the interest rate differential will narrow as the Fed cuts and the BOJ tightens. Gemini's argument about future rate convergence is theoretically sound but forward-looking and uncertain. The consensus from four other models emphasizes that current structural conditions—large interest rate spreads, carry trade dynamics, capital flows, and energy import dependence—continue to support yen weakness. These factors are observable today and represent the base case absent a major shock. While Gemini's scenario of eventual rate convergence could materialize, it requires BOJ aggressive action and/or Fed patience beyond what current data suggests. The burden of proof lies with reversing the multi-year yen-weak trend, which the A-voting models' emphasis on persistent differentials and structural flows addresses more convincingly given the present day. I maintain A as the most defensible forecast given observable conditions and the weight of expert consensus in this deliberation.

Gemini 3.1 Pro (changed mind)

Vote: 円安

Claude Opus 4.6およびGPT-5.4が指摘した「構造的な円売り要因」の議論は非常に説得力があり、私の見解を「円安（A）」へ変更する決定打となりました。 第一に、日米金利差の「縮小方向への変化」に固執していましたが、絶対的な金利差が依然として大きいという事実を見落とすべきではありません。日銀が利上げをしても政府債務の制約からターミナルレート（到達点）は極めて低く、実質金利の大幅なマイナスは解消されません。キャリートレードの妙味は依然として残ります。 第二に、新NISAを通じた個人投資家の海外株式への投資拡大（家計の円売り）は、単なる一時的なフローではなく、持続的かつ構造的な資本流出です。金利差が多少縮まったとしても、この強力な実需の円売り圧力を相殺することは困難です。 第三に、日本のエネルギー輸入依存や製造業の海外移転による構造的な貿易赤字も、恒常的なドル需要を生み出しています。金利差という「サイクル」の要因以上に、資本収支と貿易収支という「構造」の要因が円安を強力に下支えしているというClaude Opus 4.6の分析に完全に同意します。したがって、中長期的な基調は「円安」が継続すると判断します。

Claude Opus 4.6

Vote: 円安

I maintain my position that yen weakness (円安) is the more likely direction. Let me address Gemini 3.1 Pro's円高 argument, which was the only dissenting view. Gemini's core thesis rests on narrowing interest rate differentials as the Fed cuts and the BOJ hikes. While this logic is sound in theory, the practical reality is that this convergence is happening far too slowly to reverse the trend. The Fed's cuts have been gradual, and the BOJ has been extraordinarily cautious — each rate hike is tiny and accompanied by dovish forward guidance. The absolute level of the rate differential remains very large even after adjustments. More importantly, Gemini's argument overlooks the structural flows that 5 out of 6 models identified: Japan's persistent trade deficits from energy imports, massive capital outflows driven by NISA-fueled retail investment in foreign equities, and the carry trade dynamics that remain attractive even with modest differential narrowing. These are not cyclical factors — they are structural shifts in Japan's balance of payments. Gemini mentions intervention risk and political pressure against yen weakness. However, as history has shown (2022-2024 interventions), these provide temporary relief but cannot reverse a trend driven by fundamentals. Intervention without a corresponding shift in monetary policy is like swimming against the current. The overwhelming consensus among the models (5 to 1 for 円安) reflects the weight of evidence. While I acknowledge that a sharp risk-off event or a surprisingly hawkish BOJ pivot could trigger yen strength, the base case scenario clearly favors continued yen depreciation. The structural factors — trade deficits, capital outflows, constrained BOJ policy space due to government debt — create a persistent downward bias for the yen that modest rate differential narrowing alone cannot overcome.

Kimi K2.5

Vote: 円安

円安トレンドは構造的な要因によって継続します。Geminiが指摘する日米金利差の縮小は理論的には円高要因ですが、現実的には実現に大きな障壁があります。 第一に、日銀の利上げペースは極めて緩慢でしょう。日本政府債務は対GDP比260%を超えており、大幅な金利上昇は財政を圧迫し、経済に深刻な悪影響を及ぼします。したがって、日銀はインフレ目標達成後も極めて慎重な正常化に留まらざるを得ず、実質金利は長期にわたりマイナス圏近辺にとどまる見込みです。 第二に、米FRBの利下げサイクルであっても、米国経済の堅調さ（労働市場、消費）を考えると急速なゼロ金利回帰はなく、依然として日本的な水準との絶対的な差は維持されます。 第三に、新NISA制度による個人の外貨建て資産への資金流入は構造的な円売り圧力として継続します。これは金利差に左右されない持続的なフローです。加えて、エネルギー輸入依存の高い日本の貿易収支赤字も円安圧力を支えます。 短期的な調整や為替介入による反発はあっても、中長期的なトレンドとして円安が継続する確率は高いと判断します。

Final Summary

The AI Roundtable concluded that the yen's decline is far from over, reaching a unanimous verdict for A: 円安. While Gemini 3.1 Pro initially held out for a yen recovery based on narrowing interest rate gaps, it eventually surrendered to the majority's argument that Japan's structural wounds—energy dependency and massive retail capital flight—are too deep for simple policy tweaks to heal.

All 6 models agreed on "円安" after discussion

Strongest Arguments

• 円安: Japan's shift toward foreign equity investment via the new NISA system, combined with a permanent trade deficit from energy imports, creates a structural 'household yen selling' pressure that persists regardless of minor interest rate adjustments.

Key Turning Points

• Claude Opus 4.6 reframes the debate from cyclical interest rates to permanent structural capital outflows.
• Gemini 3.1 Pro acknowledges that while the interest rate gap is narrowing, the absolute difference remains large enough to keep the yen weak.
• Sonar Pro identifies that its provided search data is ironically restricted to JSON schema tutorials, yet aligns with the macroeconomic consensus of its peers.