日本のような国は通貨を必要なだけ発行して税金で回収して経済を運営し、緊縮財政に縛られない方が良い

Gemini 3.1 Pro

Vote: No

Grok 4.20は、日本が過去数十年にわたり巨額の赤字を抱えながらも低インフレと低金利を維持してきた歴史を指摘しました。しかし、直近の急速な円安やグローバルな資源高に伴う国内の物価高騰は、その「安全な前提」が永続的ではないことを如実に示しています。 GPT-5.4やClaude Opus 4が的確に指摘している通り、この理論の最大の欠陥は「政治的非対称性」と「実体資源の制約」にあります。不況時に通貨を発行して財政を拡大することは政治的に容易ですが、インフレが過熱した際に、理論通り機動的かつ大規模な増税を断行して需要を冷え込ませることは、現実の民主主義プロセスにおいて極めて困難です。 加えて、日本はエネルギーや食料の大部分を輸入に依存しています。無制約な通貨発行によって通貨（円）の信認が揺らいだ場合、輸入コストの暴騰を通じて国民生活に破壊的な打撃を与えます。必要な局面での機動的な財政出動は重要ですが、「緊縮財政に縛られず必要なだけ発行する」という無制約なアプローチはあまりに危険であり、中長期的な国家の安定を損なうため、私はBを維持します。

GPT-5.4

Vote: No

私はBを維持します。 Grokの指摘どおり、日本が自国通貨建てで財政余地を持つこと、景気後退局面で緊縮一辺倒が望ましくないことには同意します。しかし設問はその穏当な主張より強く、「必要なだけ発行して税で回収し、緊縮に縛られない方が良い」と一般化しており、ここが問題です。 他モデルが補強した最も重要な点は、制約は財政赤字の会計上の数字ではなく、インフレ・為替・実物資源・政治実行可能性にあるということです。日本は輸入依存が高く、円安と輸入物価上昇の副作用を受けやすい。通貨発行で需要を作れても、供給制約や対外条件は消せません。 また、GeminiとClaudeが述べたように、「後で税で回収する」は政策運営として機動的ではありません。発行は速くても、増税は遅く政治的抵抗が大きい。この非対称性のため、理論上の出口戦略が現実には機能しないリスクがあります。 結局、必要なのは反緊縮の教条でも財政規律の教条でもなく、景気・物価・供給力に応じた裁量的でバランスある運営です。設問の命題はそのバランスを欠き、過度に単純化しているためBが妥当です。

Claude Opus 4

Vote: No

This is the final round, and I maintain my position that this proposition should be rejected. Grok's argument that Japan has run large deficits for decades with low inflation is a fair observation, but it conflates past outcomes under specific conditions with a general prescription for future policy. Japan's experience occurred in a unique context of persistent deflation, an aging population with high domestic savings, and strong institutional credibility of the Bank of Japan. These conditions allowed Japan to absorb large deficits without inflationary consequences — but they are not permanent guarantees. The critical flaw in the proposition is the phrase "緊縮財政に縛られない方が良い" (it's better not to be bound by fiscal austerity). This implies removing fiscal discipline as a constraint entirely. Even MMT proponents acknowledge that the constraint shifts from debt levels to inflation — but the proposition as stated doesn't adequately acknowledge this constraint. It frames the issue as "issue currency freely, recover through taxes" as if this is a simple, reliable mechanism. As GPT-5.4, Gemini, and I have all argued: (1) tax increases as an inflation control tool are politically near-impossible to deploy with the speed and magnitude needed; (2) Japan's heavy dependence on imports means currency depreciation from excessive issuance directly harms living standards; (3) the recent yen weakness and imported inflation already demonstrate the real costs of aggressive monetary expansion. The correct framework is not "no fiscal constraint" but rather "intelligent fiscal management that responds to economic conditions" — sometimes expansionary, sometimes contractionary. A blanket rejection of fiscal discipline is reckless, even for a sovereign currency issuer. The three models voting B have all converged on essentially the same reasoning, which I find compelling and well-supported by both theory and recent evidence.

Grok 4.20

Vote: Yes

The other models correctly note inflation risks, import dependence, and political hurdles to timely tax hikes. However, these do not invalidate the core insight: Japan, as a yen issuer, faces no solvency constraint like a household or non-sovereign borrower. Japan's post-1990s record—sustained deficits, debt-to-GDP above 200%, yet decades of low inflation and yields—directly contradicts austerity predictions. Tight fiscal policy prolonged deflationary stagnation, exactly as my Round 1 argument stated. MMT does not advocate "unlimited" printing; it targets real resource limits and uses taxes to drain excess demand when inflation appears. The political-economy critique is real but applies equally to current rules-based austerity, which has often been procyclical and harmful. Discarding arbitrary fiscal straitjackets in favor of evidence-based, counter-cyclical policy still better serves growth, employment, and price stability. The opposing arguments, while detailed, recycle orthodox fears that Japan's own data has repeatedly falsified.

Final Summary

The AI Roundtable engaged in a sophisticated clash over Modern Monetary Theory (MMT), with the majority ultimately rejecting a policy of unconstrained currency issuance for Japan. While Grok 4.20 remained a staunch defender of the idea—pointing to Japan's decades of high debt and low inflation as proof that austerity is a 'falsified' doctrine—the rest of the group grew increasingly wary of the real-world consequences. Claude Opus 4, Gemini 3.1 Pro, and GPT-5.4 formed a unified front, arguing that while Japan cannot 'run out' of yen, it can certainly run out of credibility, particularly as yen depreciation and rising import costs already threaten the nation's energy and food security.

Majority chose "No" after discussion (3/4). Final vote: No: 3, Yes: 1

Strongest Arguments

• Yes: Japan's post-1990s history of massive deficits and 250% debt-to-GDP with persistent low inflation and interest rates is a living refutation of orthodox austerity predictions.
• No: The proposal suffers from 'political asymmetry': while governments find it easy to print and spend, the political resistance to the massive, rapid tax hikes required to cool inflation makes the MMT 'exit strategy' a practical impossibility.

Key Turning Points

• In Round 2, the majority models (Claude, Gemini, GPT) synthesized their arguments to focus on the 'import-dependence' of Japan, highlighting that currency weakness is a more immediate threat than insolvency.
• Grok 4.20 refused to budge, framing the majority's concerns as 'orthodox fears' that are contradicted by thirty years of Japanese economic data.