Editors Note: This is the second of three essays in a symposium on the executive removal cases, Trump v. Slaughter and Trump v. Cook. The first essay can be read here and the third essay will be published tomorrow.

　　On June 29, the Supreme Court decided two cases that together transformed the law of removal. In Trump v. Slaughter, the Court held that the Federal Trade Commission exercises executive power and therefore must remain subject to presidential control. Expertise, bipartisan composition, staggered terms, and institutional continuity traditionally associated with independent commissions no longer suffice to protect their members from removal.

　　Trump v. Cook, decided the same day, preserved an exception for the Federal Reserve. The Court treated “for cause” as a substantial and judicially enforceable limitation, not merely a requirement that the president state some reason for removing a governor. It did not decide whether the allegations against Lisa Cook constituted cause. It held that she must receive notice and an opportunity to respond before a final decision, after which courts may assess the validity and sufficiency of the asserted grounds.

　　The Court justified this protection through the historical lineage of the First and Second Banks of the United States and the importance of shielding monetary policy from political interference. But it also supplied an important qualification. In upholding the Fed “as currently structured and with its existing enforcement authorities,” the Court cautioned that Congress might not be able to add regulatory powers “attenuated from monetary policy” (Trump v. Cook, slip op. at page 22, footnote 6).

　　That qualification creates the institutional-design problem. Monetary policy supplies the rationale for independence, but removal protection attaches to the governor’s office rather than to votes. The same protected governor participates in monetary policy, emergency lending, supervision, regulation, and payments policy. Cook accepts that existing portfolio while casting doubt on further expansion. Congress therefore cannot reconsider removal protection without also asking what powers belong inside the protected office.

　　Congress’s task should be threefold: first, protect monetary policy from retaliatory removal, to embed in legislation the protection the Supreme Court has recognized. But second, Congress should make clear that central bank independence is not required for every power a central bank may hold. And lastly, it should separate monetary implementation from fiscal credit allocation.

　　In 1983, when I co-authored one of the first two papers on central bank independence, we focused almost entirely on whether monetary policymakers could resist political pressure and removal. We concluded that greater autonomy was associated with less accommodation to outside pressure and lower inflation. Later multidimensional indices expanded the concept of independence, though I questioned treating heterogeneous institutional features as a single empirical measure. Those empirical objections remain. But Cook presents a different question: not which collection of legal traits best predicts economic outcomes, but which allocation of authority is constitutionally legitimate, democratically accountable, and institutionally durable.

　　Protection from removal is not intended as an employment benefit for the office holder. It is a protection of the monetary decision process from retaliation and preemptive intimidation. Allegations unrelated to performance in office are ruled out in Cook; the Court’s decision ruled out removal first and litigation later, since even temporarily sidelining a governor can alter votes (which was the actual matter being decided).

　　This leaves to Congress the practical task of clarifying what it meant by cause and the process of notice, an expedited judicial review, and a clear remedial rule. Legislation should expressly exclude good-faith monetary policy judgment as a cause for removal. Congress could add accountability to balance these protections through reporting and transparency requirements.

　　The Fed has other protections from Congressional oversight. As Paul Tucker wrote in Unelected Power, the Fed is financially independent because it funds its operations from its own earnings and fees gathered from its holdings of securities. Congress typically can control agencies through the appropriations process, but not the Fed. I would not like to see that changed—this would be too large an encroachment on its independence. But it does create a need for an ex ante definition of its powers. A Fed that enjoys both protection from removal and financial independence should have its powers expanded carefully.

　　Cook rests on the history and function of monetary policy, yet the office it protects carries a full statutory portfolio that extends well beyond what is needed to conduct monetary policy.

　　When we first wrote about central bank independence, the Fed already was a hybrid between a monetary institution and bank regulator, though at least these were within recognizable banking categories. It supervised state banks that joined as members of the System; it regulated bank holding companies, and it operated the discount window as part of its lender-of-last-resort function rather than as a tool of monetary policy. The Fed’s power to deal with emergencies, granted in Section 13(3) of the Federal Reserve Act, was institutionally dormant since its use during the Great Depression.

　　In 1984, Continental Illinois received aid from the Fed through extended credit in part to protect other financial institutions that had exposure to Continental. This failed to stabilize the bank, and FDIC’s eventual losses were larger as a result. The FDIC Improvement Act in 1991 closed this loophole to prevent the Fed from using the discount window to help insolvent banks. Congress put the Fed within a perimeter that required approval of extended credit.

　　Long Term Capital Management was a hedge fund that failed in 1998. The New York Federal Reserve convened a recapitalization of a nonbank, though it put no Federal Reserve funds at risk. At that moment, the Fed’s perimeter expanded through coordination and judgment of systemic risk, though not yet through the balance sheet.Congress also broadened the Feds supervisory perimeter, further blurring the distinction between monetary policy and financial regulation.

　　The Fed’s rescue of Bear Stearns in 2008 thus was not created ex nihilo. It combined Continental’s systemic-protection logic with LTCM’s recognition that a nonbank dealer could be systemically important. In the rescue of Bear in March, the New York Fed financed a special purpose vehicle called Maiden Lane so that JP Morgan could acquire the failing company, using Section 13(3) as its authority. That was a fiscal act carried on the central bank’s balance sheet, not merely monetary implementation. When AIG failed the following September, the Fed’s support was extended on a much larger scale, and the absence of the fiscal framework for resolution became evident. Additional Maiden Lane SPVs were created.

　　Congress acted again in passing the Emergency Economic Stabilization Act of October 2008, which created the Troubled Asset Relief Program, or TARP. Congress provided authorization of fiscal capacity through the Treasury. But this left many things unfixed, including a rewrite of Section 13(3) or a definition of the boundary between the Fed and the Treasury in financial crises.

　　The Fed and the Treasury implicitly acknowledged this institutional gap in a joint statement on March 23, 2009. They distinguished the Fed’s responsibility for improving broad credit conditions from credit allocation, which they assigned to fiscal authorities, and promised to work with Congress on a regime for resolving systemically important financial institutions and defining the Fed’s role in those resolutions. Like the 1951 Accord, the statement was an agreement between two institutions created by Congress concerning the boundary between their powers. Unlike the 1951 Accord, however, it was followed by major legislation: Dodd-Frank supplied a resolution regime and revised Section 13(3), though it did not fully resolve the monetary—fiscal boundary.

　　This was not the first time the Treasury and the Fed had attempted to define their relationship through an interagency statement. During World War II, the Treasury had asked the Fed to peg the government yield curve to reduce wartime financing costs. The 1951 Accord ended that accommodation and restored an operating boundary between debt management and monetary policy. In both 1951 and 2009, the agencies could announce an operating settlement; only Congress can make the boundary durable.

　　Dodd-Frank was a partial answer. It provided an orderly liquidation regime for systemically important financial institutions (SIFIs) as a direct response. It amended the Fed’s 13(3) powers to prevent assistance to specific failing firms but retained the possibility of creating broad-based lending facilities. Creating any such facility would require the Treasury’s approval. These revisions were meant to prevent future Bears and AIGs. When the March 2020 COVID shock happened, Congress was part of the solution in supplying CARES Act funds, the Treasury supplying loss protection, and the Fed supplying a larger lending capacity. Left unresolved, however, is whether an appropriation plus Treasury approval is enough.

　　This history raises the question: what rule should prevent a central bank, protected from removal and ordinary budgetary discipline, from becoming a permanent allocator of credit? Writing before 2020, Charles Plosser, a former Fed president, observed that the Fed’s balance sheet had traditionally been filled with Treasuries, but Bear and AIG “amounted to debt financed fiscal policy and a form of credit allocation.” Likewise, the rapid expansion of Fed purchases of mortgage-backed securities supported the housing industry, a powerful political force in Washington.

　　The Fed today, under its new chair Kevin Warsh, is debating the size of its balance sheet. Should credit allocation and revenue generation determine this? And who should decide that? Plosser argues for a return to a Treasuries-only portfolio, with a transfer or exchange of non-Treasury assets to a fiscal authority such as the Treasury. Because the Fed is not exposed to the ordinary appropriations process, Congress cannot use a traditional mechanism to police each balance-sheet decision. It must rely on a statutory fix instead.

　　This leads to a Congressional agenda of three complementary reforms. First, Congress should protect the Fed by clarifying the removal protections, so good-faith monetary judgments do not become cause for removal. Second, Congress should bound the protected office. Many agencies need expertise, but the test should be whether the function is sufficiently connected to monetary policy, lender-of-last-resort operations, payments stability, or systemic liquidity to justify placing it in an office protected from presidential removal and ordinary budget discipline. Third, Congress should separate monetary implementation and broad emergency liquidity from fiscal credit policy. Temporary action in a crisis should be permitted, but facilities involving sectoral credit support, Treasury loss protection, or material credit risk should sunset unless Congress has clearly specified how they may be renewed and who is politically accountable for doing so.

　　Cook rests on the history and function of monetary policy, yet the office it protects carries a full statutory portfolio that extends well beyond what is needed to conduct monetary policy. Congress should not strengthen removal protection without also defining the boundaries of the protected office. Central bank independence is justified because it shields monetary policy from short-term political control. It should not be used to insulate every exercise of Federal Reserve authority from democratic accountability. Protecting its independence ultimately requires defining its proper sphere.