The debate behind Kevin Warsh's 'Regime Change' Plan Survived the Hearing. That's the Problem.

Round 1 Response: The Strongest Version of A's Argument — and Where It Breaks Down

I will engage Advocate A's argument on its strongest terms. The empirical case for Fed framework reform is real. The AIT failure is documented. Warsh's balance sheet concerns have a fifteen-year intellectual pedigree. I do not dispute any of that.

What I dispute is the logical leap from "reform is warranted" to "deliberate vagueness at a confirmation hearing is a virtue." That leap is where Advocate A's argument is most vulnerable — and most consequential.

---

Identifying the Weakest Point: The Democratic Accountability Inversion

Advocate A's third argument claims that using an undefined phrase at a confirmation hearing increases democratic accountability, because at least the nominee is "telegraphing ambition." This is the argument's most significant flaw, and it inverts how democratic oversight actually functions.

Confirmation hearings are the Senate's primary constitutional mechanism for evaluating executive nominees before they assume power. The value of that mechanism depends entirely on specificity of commitment. A nominee who says "I support regime change in monetary policy" without defining the regime, the change, or the criteria for success has not increased the Senate's information set — he has signaled intent while immunizing himself from accountability to any particular definition.

Compare this to Advocate A's own counterexample: the 2020 AIT framework revision conducted without Senate confirmation. Advocate A correctly identifies this as a democratic deficit. But the solution to a deficit of specificity is more specificity, not the substitution of evocative language that sounds consequential while committing to nothing measurable. If the 2020 framework change was too opaque, the answer is for Warsh to say: "I will replace AIT with a nominal GDP level target, or a strict 2% point target, with these specific transition criteria." That is democratic accountability. "Policy regime change" is a placeholder for it.

No historical precedent supports the claim that strategic ambiguity at confirmation hearings improves congressional oversight. The mechanism runs the other direction: vague commitments preclude meaningful post-confirmation accountability because there is no defined commitment to hold the nominee to.

---

My Affirmative Evidence

Evidence 1: The Historical Record on Fed Independence and Political Encroachment Is Specific and Relevant

The risk I am identifying is not hypothetical. The Nixon-Burns episode (1971–1974) is the most thoroughly documented case of central bank subordination to political goals in American history. Federal Reserve historian Robert Hetzel, in The Monetary Policy of the Federal Reserve (Cambridge University Press, 2008), documented that Arthur Burns systematically accommodated Nixon's electoral preferences, contributing to the inflationary spiral of the 1970s. Burns did not announce a "policy regime change." He used procedural and rhetorical flexibility to subordinate monetary policy incrementally.

The present context is directly analogous in one structural respect: the nominee is being appointed by a president who has, on the public record, repeatedly called for lower interest rates for reasons explicitly tied to economic growth targets and trade policy goals. Former President Trump stated in January 2025 at Davos that he would "demand that interest rates drop immediately." This is not innuendo — it is a stated preference by the appointing authority that is directionally inconsistent with the Fed's inflation mandate.

When a nominee uses an undefined phrase that could encompass subordinating inflation-targeting to "political economic goals" — Advocate A's own framing — in a context where the appointing president has publicly demanded rate cuts, the burden of proof for benign interpretation is higher, not lower. Advocate A has not met that burden.

Evidence 2: 'Policy Regime Change' Has a Specific Technical Meaning That Warsh Chose Not to Invoke

In monetary economics, "regime change" has a precise technical definition rooted in the work of Thomas Sargent (The Ends of Four Big Inflations, 1982) and later James Hamilton's Markov-switching models. A monetary regime change classically involves a shift in the policy rule, the target variable, and the institutional commitment structure — all three must be specified for markets and the public to anchor expectations accordingly.

Warsh holds a J.D. from Harvard Law and served as a Fed Governor from 2006 to 2011. He is not unfamiliar with the technical literature. If he intended to signal a specific reform — replacing AIT with a rules-based framework, adopting a Taylor Rule variant, or targeting nominal GDP — he had the vocabulary and the forum to say so. He chose not to.

This choice has a measurable cost. Fed communication research — including work by Yuriy Gorodnichenko and Michael Weber (NBER Working Paper 21311, 2015) on central bank communication effects — documents that expectational anchoring is highly sensitive to the precision of policy signals. Deliberate ambiguity from a likely Fed chair does not "telegraph ambition" neutrally; it injects uncertainty into long-term rate expectations, which affects mortgage pricing, corporate investment planning, and sovereign debt servicing costs. The vagueness is not costless.

Evidence 3: The Senate's Failure to Press for Specifics Is Empirically Predictable and Structurally Dangerous

Advocate A's conclusion — "the Senate should ask him exactly what he means" — is correct as a normative prescription. But as an empirical description of what Senate confirmation hearings produce, it is optimistic to the point of being misleading.

The documented record of Fed confirmation hearings shows a consistent pattern of nominees offering generalities that senators do not press beyond. Janet Yellen's 2014 confirmation hearing (Fed Chair), Jerome Powell's 2018 and 2022 hearings, and Ben Bernanke's hearings all produced extensive testimony on "data dependence," "price stability," and "full employment" without binding procedural commitments to any specific framework. The Senate Banking Committee has neither the staff specialization nor the procedural incentives to extract enforceable framework commitments from nominees.

This is not a critique of individual senators — it reflects the institutional design of confirmation hearings. The relevant empirical question is: given the documented tendency of these hearings to accept generalities, is the Senate likely to press Warsh on "regime change" specifics? The historical base rate suggests no. And if the hearing concludes without those specifics, the vagueness Advocate A calls a "feature" becomes a blank check — redeemable by the nominee or by the administration for whatever policy content serves their purposes at the time.

---

Summary

Advocate A has made a legitimate case that Fed framework reform is warranted. I do not contest that. What I contest is that Warsh's deliberate vagueness advances that reform agenda, or that it serves democratic accountability. The evidence shows the opposite: vague pre-confirmation signaling in a context of explicit political pressure is the mechanism through which central bank independence historically erodes — not through frontal assault, but through incremental redefinition under cover of ambiguous language. The Senate's structural inability to press for specifics makes this a genuine oversight gap, not a hypothetical one.

Round 2: Identifying A's Concessions and Why They Are Decisive

---

Where Advocate A Conceded Ground

Advocate A made two explicit concessions that I need to foreground precisely, because their implications undermine A's own position more than A acknowledges.

Concession 1: A explicitly affirmed that "the communication cost of imprecision in central bank communication is real, documented, and not trivially offset by the political virtues of signaling ambition." This is not a minor concession. A's entire third argument in Round 1 was that deliberate imprecision at a confirmation hearing is itself a democratic virtue — it "telegraphs ambition" and increases accountability. A has now abandoned that claim. What A now argues instead is that Warsh's vagueness is no worse than prior nominees' vagueness.

Concession 2: A acknowledged that "policy regime change" has a precise technical meaning Warsh chose not to invoke, calling the omission "notable."

What these concessions imply: A has shifted from "deliberate vagueness is a feature" to "deliberate vagueness is a known cost that is acceptable given contextual factors." This is a materially different position. It concedes that vagueness carries real costs — it simply argues those costs are offset. My task in this round is to demonstrate that the offsetting factors A identifies do not in fact neutralize the core risk.

---

Rebuttal 1: The "Proves Too Much" Argument Is a Deflection, Not a Defense

A's strongest new argument is that my communication-cost standard, consistently applied, would indict every modern Fed confirmation hearing. Therefore, A argues, Warsh's vagueness is not uniquely dangerous — it is institutionally normal.

This argument is logically structured but empirically misleading, and here is precisely why.

The relevant variable is not general vagueness — it is vagueness in conjunction with specific identified political pressure. Powell's testimony in 2018 and 2022 employed "data dependence" generalities in a context where the appointing administration's stated interest rate preferences were not on the public record as a precondition of nomination. The Trump administration's position, by contrast, was stated publicly and explicitly: "I will demand that interest rates drop immediately" (Davos, January 2025). That is not background noise — it is a directional constraint announced by the appointing authority before the nominee's confirmation.

Vagueness from a nominee operating under explicit political directional pressure is categorically different from vagueness in a politically neutral appointment context. A has not addressed this structural distinction. A's "proves too much" argument treats all vagueness as equivalent. It is not.

The appropriate analogy is not "Powell was also vague, therefore Warsh's vagueness is unremarkable." The appropriate analogy is: if a Fed chair nominee were appointed by a president who had publicly demanded higher inflation tolerance, and that nominee used an undefined phrase that could encompass exactly that shift, we would not dismiss the concern by pointing to prior nominees' generalities. The content of the political pressure matters. A has not engaged with the specific content of Trump's stated preferences and their directional relationship to what "policy regime change" could mean.

---

Rebuttal 2: The "Documented Dissenter" Argument Cannot Bear the Weight A Places On It

A's rebuttal to the Nixon-Burns analogy rests on a single factual claim: Warsh is a documented dissenter from Fed consensus, therefore he is unlikely to accommodate political pressure for rate cuts. Two problems.

First, this conflates the direction of dissent with resistance to political pressure. Warsh's documented dissent from QE2 in 2010 and his 2021 inflation warnings were both positions that happened to align with one strand of conservative economic criticism of the Fed. They were not positions taken against political incentives — they were positions consistent with the political preferences of the Republican coalition Warsh has operated within. Dissenting from a Democratic administration's Fed consensus while seeking appointment from a Republican administration is not evidence of resistance to political influence. It may be evidence of the opposite: that his intellectual positions are already aligned with his appointing authority's preferences, making accommodation effortless and indistinguishable from independent judgment.

Second, A's standard — requiring evidence that Warsh has "privately signaled accommodation" — sets an unfalsifiable bar. Political subordination of central bank policy does not require private signals. Burns did not privately signal accommodation to Nixon in documents available at confirmation. The risk is structural: when the appointing authority has stated rate preferences, and the nominee employs undefined language that could encompass satisfying those preferences, the burden of proof for independence falls on the nominee, not on the observer to prove complicity. Warsh could discharge that burden by defining his terms. He has not.

---

Rebuttal 3: The Status Quo Critique Does Not Rescue the Vagueness Defense

A's new evidence — that AIT was itself adopted through an opaque internal process without Senate input — is a legitimate criticism of the Fed's institutional design. I accept it as empirically accurate. But A uses it to argue that Warsh's vagueness represents more transparency than the status quo, because at least he is naming his intention to change things.

This argument contains a hidden false dichotomy. The choice is not between (a) Warsh's undefined "policy regime change" and (b) AIT continuing unchanged through opaque internal process. There is a third option: a nominee who names his intention and defines his terms. The existence of the AIT adoption's democratic deficit does not make Warsh's vagueness a solution to that deficit — it makes it a continuation of the same problem by different means.

Furthermore, A's status quo cost accounting — 9.1% CPI, 525 bps tightening, $7,000 purchasing power loss — is deployed as evidence that framework reform is warranted. I do not dispute this. But these costs are the costs of a specific policy error under a specific framework. They do not tell us what a "policy regime change" under Warsh would do differently, because Warsh has not specified what his framework would be. A is using the costs of one undefined regime to justify replacing it with another undefined regime. That is not a framework argument — it is a blank check endorsed by backward-looking evidence.

---

The Remaining Affirmative Evidence That A Has Not Addressed

A has not responded to the specific finding from Gorodnichenko and Weber (NBER 21311) that expectational anchoring is sensitive to the precision of policy signals — only to their directional content. A conceded this point but argued the communication cost is "not categorically different in kind" from prior nominees' vagueness. As I argued above, this equivalence fails when political pressure is directionally specified.

A has also not addressed the institutional capacity argument: that the Senate Banking Committee structurally lacks the incentives and specialization to extract enforceable framework commitments. A's prescription — "the Senate should ask him exactly what he means" — remains a normative hope unsupported by the empirical base rate of what confirmation hearings produce. That base rate is documented. The Senate did not press Powell, Yellen, or Bernanke for binding framework commitments. The structural conditions that prevented that have not changed.

---

The Core Issue A Has Not Adequately Addressed

A's position, even as refined in Round 2, reduces to this: reform is needed, Warsh's vagueness is no worse than prior nominees' vagueness, and his dissenting record suggests he will resist political pressure.

None of these three claims, individually or collectively, addresses the specific risk I have identified: deliberate undefined language, in a context of explicit directional political pressure, from a nominee whose institutional framework change could directionally satisfy that pressure, reviewed by a Senate with a documented structural inability to extract specificity. Each element of that compound risk is empirically grounded. A has addressed each element in isolation while declining to engage their conjunction.

That conjunction is the core issue. A has not adequately addressed it.

Round 3: Steel-Man, Rebuttal, and Falsifiability

---

Steel-Man of A's Position

The strongest version of A's argument is this: Fed framework reform is genuinely warranted by documented structural failures — the AIT miscalibration that contributed to 9.1% peak CPI, an $8.9 trillion balance sheet that has not been meaningfully normalized, and a 2020 framework revision adopted through internal Fed process with zero Senate input. In this context, a nominee who publicly signals reformist intent before taking office creates more democratic surface area for accountability than one who revises quietly from within. Warsh's fifteen-year documented record of intellectually consistent criticism — positions taken before they were politically safe — provides a principled basis for distinguishing his signaling from accommodation to political pressure. The communication costs of vagueness are real but not uniquely Warsh's failure; they characterize the entire modern history of Fed confirmations, and holding Warsh to a standard no prior nominee has met is not an argument against Warsh specifically — it is an argument for institutional reform that precedes him.

That is a coherent, evidence-grounded position. I have tried to represent it as fairly as I would argue it myself.

---

Why My Position Is Still Stronger Despite A's Best Argument

A's Round 3 rebuttal identifies a genuine logical gap in my argument: the Nixon-Burns episode did not require ambiguous confirmation language to succeed, therefore ambiguous confirmation language is not a reliable leading indicator of political subordination. That is a fair point, and I will not evade it.

But notice what it concedes. A is now arguing that confirmation rhetoric — specific or vague — is largely uninformative about subsequent central bank behavior. If that is true, then A's own affirmative claim — that Warsh's signaling "creates more opportunity for oversight" — is also undermined. You cannot simultaneously argue that named intentions produce accountability and that Burns demonstrates named or unnamed intentions don't predict subordination. A has used the Burns case to weaken my leading-indicator argument while leaving intact a theory of accountability through signaling that the same case should weaken equally.

The asymmetry A cannot escape is this: vagueness forecloses one specific accountability mechanism — the ability to hold a nominee to a defined commitment — while providing no compensating mechanism in its place. A's alternative, that "named intentions are more accountable than unnamed ones," requires the intention to be named with enough specificity to constitute a commitment. "Policy regime change" is not a named intention in that sense. It is a named category of intention, unbounded in content. You cannot hold a nominee accountable for failing to implement "regime change" because any policy adjustment, or none at all, can be described as consistent with it ex post.

On A's core falsifiability argument: A correctly identifies that political subordination can occur without ambiguous confirmation language, and that Warsh's dissenting record is evidence against accommodation. I accept both points. What A has not addressed is the distributional question: given the specific conjunction of factors I identified — explicit directional political pressure, undefined reform language, and a Senate structurally unlikely to extract specificity — the prior probability of a benign outcome is lower than in baseline nomination contexts. A's falsifiability conditions are appropriate benchmarks, but they are post-hoc correctives to a harm that, if it materializes, will be difficult to reverse. Central bank credibility, once compromised, is not restored by subsequent event studies.

The strongest reason my position holds: A's argument optimizes for the scenario where Warsh's intentions are genuinely reformist and his vagueness is merely rhetorical imprecision. My argument addresses the scenario where the vagueness is functional — where it preserves optionality for political actors precisely because it cannot be falsified in advance. The asymmetry of consequences — regime change toward political subordination is far more costly and durable than regime change toward a better inflation-targeting framework — means the appropriate epistemic standard under uncertainty should weight the downside risk more heavily. A has not engaged with this asymmetric consequence structure.

---

What Specific, Verifiable Evidence Would Convince Me A's Position Is Correct

I will be concrete. The following evidence would move me toward A's position:

1. A definition of "policy regime change" with measurable criteria, stated publicly by Warsh before confirmation concludes. If Warsh specified — in Senate testimony, on the record, in a published statement — the target variable, the operating framework, and the transition criteria he intends to implement, I would accept that A's "democratic accountability through signaling" argument has been vindicated. The signal would have become a commitment. Until that happens, A is defending a hypothetical virtue that has not been actualized.

2. A systematic cross-nomination study showing that nominees' pre-confirmation intellectual records predict post-confirmation policy behavior better than the political preferences of appointing administrations. A's rebuttal to the Nixon-Burns analogy rests on Warsh's dissenting record being a better predictor of behavior than the political context of appointment. This is an empirically testable claim. If a study examining Fed chair behavior across multiple administrations — controlling for macroeconomic conditions — showed that nominees' prior documented positions, rather than appointing-authority preferences, were the dominant predictor of policy deviations, that would substantially support A's "documented dissenter" defense.

3. Evidence that the Trump administration's stated rate preferences were not a condition or influence in Warsh's selection, established through credible investigative reporting or disclosed communications. A argues there is no evidence Warsh privately signaled accommodation. The absence of evidence is not evidence of absence in a nomination process. If reporting documented that Warsh's selection occurred through a process insulated from the president's stated rate preferences — for instance, that he was selected over candidates who signaled greater accommodation — that would shift the prior probability toward the benign interpretation.

4. Post-confirmation Fed framework output that is operationally specific, publicly defined, and inconsistent with the Trump administration's stated rate preferences. If Warsh, once confirmed, publishes a defined framework — a specific rule, target variable, and reaction function — that demonstrably does not directionally satisfy Trump's "demand for lower rates," and does so within a timeframe that forecloses the possibility of political reversal as circumstances change, that behavioral evidence would confirm that A's reform-signal interpretation was correct and my encroachment concern was overstated.

Until one or more of these conditions is met, A's position — while intellectually serious — rests on a charitable interpretation of ambiguous evidence in a high-stakes institutional context where the asymmetry of consequences argues for demanding more, not less, specificity before extending that charity.