In-Depth Breakdown of the Federal Reserve’s Major “Balance Sheet Reduction” Paper: How Much to Shrink, How to Shrink, and to What Extent Will It Affect Things?

On Tuesday evening at 10 p.m. in Taiwan, the U.S. Senate Banking Committee will hold a hearing on Kevin Waugh’s nomination for Federal Reserve Chair. This is Waugh’s first formal appearance at the Capitol to systematically articulate his monetary policy stance. Notably, Waugh has long been critical of the Fed’s massive balance sheet, and this hearing may serve as an important platform for him to express related views.

In fact, since the end of 2025, the direction of the Fed’s balance sheet has been a core topic of high concern in global financial markets. Against this backdrop, Fed Governor Stephen Milan, together with three Federal Reserve economists, recently released a working paper titled “A User’s Guide to Shrinking the Federal Reserve’s Balance Sheet,” and will use a keynote speech at the Miami Economic Club on March 26, 2026, to systematically explain the strategic logic and potential pathways for Fed balance sheet reduction.

The core value of this paper lies in challenging conventional market perceptions. Previously, the market generally believed that “the ceiling for Fed balance sheet reduction is when reserves are exhausted.” However, the paper points out that reserve demand can itself be shaped by policy—through a series of regulatory and operational adjustments, the Fed can significantly shrink its balance sheet while maintaining a “sufficient reserves” framework.

In response, CITIC Securities’ research team has provided an in-depth interpretation. They judge that options such as relaxing the LCR standards, reforming SRP, and upgrading Fedwire are somewhat feasible; but proposals like layered reserve requirements, reforming TGA and foreign reverse repo pools are more idealistic. Overall, the process of balance sheet reduction is unlikely to alter the underlying logic of global central banks’ gold purchases. CITIC Securities maintains their view that the Fed will cut interest rates by 25 basis points in the second half of this year.

Why Reduce the Balance Sheet: Milan’s List of Reasons

In his Miami speech, Milan straightforwardly presented multiple reasons for reducing the Fed’s balance sheet.

First, to reduce market distortions. The Fed’s large balance sheet exerts unnecessary interference on the capital markets, exacerbating disintermediation issues. Minimizing the Fed’s “footprint” in the market is a fundamental requirement for maintaining proper price discovery.

Second, to control financial risks. Large asset holdings mean greater exposure to market value losses and increased volatility in remittances to the Treasury. Recently, the Fed has been under pressure from unrealized losses due to holding long-duration securities, which is an issue that can no longer be ignored.

Third, to safeguard monetary-fiscal boundaries. The enormous balance sheet objectively allows the Fed to intervene in credit resource allocation, blurring the line between monetary and fiscal policy. Additionally, paying large amounts of interest on reserves has been viewed by some Congress members as a hidden subsidy to financial institutions.

Fourth, to retain policy ammunition. If the next zero-lower bound crisis occurs, the Fed will need to expand its balance sheet to provide room for easing. Shrinking the balance sheet now to a reasonable size preserves necessary space for future policy adjustments.

Milan admits that the public generally believes that large-scale balance sheet reduction is “impossible.” But his judgment is quite different: “Balance sheet reduction is a solvable challenge; those who outright deny it simply lack imagination.”

Key Diagnosis: Demand, Not Supply, Is the Bottleneck for Balance Sheet Reduction

To understand this discussion, one must first clarify a long-misinterpreted logical structure.

The traditional view holds that the constraint on Fed balance sheet reduction comes from “reserve supply hitting the demand curve’s steep segment”—once supply tightens to a critical point, overnight rates will spiral out of control. Therefore, the Fed can only passively stop shrinking once reserves become “scarce.” The September 2019 repo market turmoil was a real-world example of this logic.

The breakthrough in the paper is shifting the perspective from the “supply side” to the “demand side.” It points out that reserve demand is not an exogenous constraint naturally determined by payment settlement activities, but is artificially elevated by regulatory rules, supervisory practices, and the Fed’s own operational frameworks. Milan refers to this phenomenon as “regulatory dominance” over the Fed’s balance sheet.

Specifically, three mechanisms jointly push up reserve demand:

1. The interest rate differential makes reserves a “risk-free earning asset.” After the Fed started paying interest on reserves (IORB) in 2008, reserves transformed from mere settlement necessities into assets competing with Treasury bills. Historically, periods when the reserve interest rate (IORB) exceeded the yields on 1- or 3-month Treasury bills saw banks prefer to hold reserves for risk-adjusted returns.

2. Multiple liquidity regulation overlays create a “ratchet effect.” Rules such as LCR (Liquidity Coverage Ratio), ILST (Internal Liquidity Stress Test), RLEN (Disposal Liquidity Assumption), NSFR (Net Stable Funding Ratio), SLR (Supplementary Leverage Ratio) intertwine, forming a “rob Peter to pay Paul” dilemma—changing one rule immediately makes another the new binding constraint.

3. The long-standing “stigma” of discount window rates. High discount window rates, historical associations with “problem banks,” and the risk of information disclosure and regulatory scrutiny discourage banks from using these tools, leading them to hoard reserves instead. This stigmatization also extends to standing repo facilities (SRP).

This diagnosis implies a fundamental policy path: instead of waiting for reserves to become scarce again, the boundary between “scarcity” and “adequacy” can be lowered, allowing the sufficient reserves framework to operate normally at a smaller balance sheet size.

How Much Can Be Reduced: Quantitative Estimate of $1.2 Trillion to $2.1 Trillion

Using the Fed’s H.4.1 report data as of March 11, 2026, the total assets were approximately $6.646 trillion. The liability structure breakdown is as follows: reserves about $3.073 trillion, currency in circulation $2.39 trillion, the Treasury General Account (TGA) about $806 billion, and foreign reverse repo pools about $325 billion.

The paper estimates the impact of 15 policy options across two main directions, but importantly, it rejects simple summation due to correlations and substitutability among policies. Using a Monte Carlo aggregation under the OMB A-4 framework, the confidence intervals are:

Median reserve demand reduction within 95% confidence: $825 billion to $1.75 trillion, approximately $1.287 trillion.
Total balance sheet reduction: $1.15 trillion to $2.125 trillion, approximately $1.637 trillion.

Milan compares these ranges with historical reference points:

• 15% of GDP: the level of assets and liabilities after the first round of QE in 2009, when the banking system was still functioning normally;
• 18% of GDP (2012 or 2019 levels): reflecting the true liquidity needs of banks after Basel reforms and Dodd-Frank clarifications.

Currently, the Fed’s balance sheet is about 21% of GDP. Based on the median estimate in the paper, if reforms proceed smoothly, the balance sheet could fall back to levels close to 2012 or 2019. As for returning below 10% of GDP before the crisis—Milan explicitly states: “Unrealistic and unnecessary.”

How to Reduce: “Menu-Style” Analysis of 15 Options

The paper categorizes the 15 policy tools into two main groups, providing effect ranges and implementation assumptions for each.

First group: Lowering the equilibrium reserve demand

(1) Supervisory reforms

• LCR reform (Liquidity Coverage Ratio): The core measure allows banks to include financing capacity from non-HQLA loans pledged at the discount window, with an upper limit. The estimated impact on reserve demand ranges from $50 billion to $450 billion. The paper also warns that if only LCR is changed, NSFR might immediately become the new binding constraint, requiring coordinated consideration.
• ILST and RLEN (Disposal Liquidity Assumption): If regulators approve discount window capacity and short-term liquidity sources, reforms here could reduce reserve demand by $50 billion to $200 billion; extending the discount window duration (RLEN) could further reduce demand by $0 to $100 billion.

(2) Supervisory approach

If banks hold excess reserves to meet examiners’ preferences (i.e., T-bills and reserves are not “equivalent” at the supervisory level), adjusting the supervisory stance could reduce reserve demand by $25 billion to $50 billion. This is a reform achievable through cultural change without legal amendments, but still challenging.

(3) Lower reserve holding returns

Allow the effective federal funds rate (EFFR) to exceed IORB, breaking the current situation where EFFR is persistently below IORB. Using Lopez-Salido and Vissing-Jorgensen (2025) framework, an “EFFR - IORB = +2bp” (close to September 2019 stress levels) could reduce reserve demand by $150 billion to $550 billion.

However, this path has clear costs: increased volatility in overnight and repo rates, and potential demand increase if markets preemptively hoard. Supporting mechanisms like SRP and temporary open market operations (TOMO) would be necessary.

(4) Enhance attractiveness of alternative assets

Upgrading Fedwire, improving Treasury market liquidity, promoting central clearing, etc., aim to make assets like Treasuries more attractive to banks, closer to reserves. These measures also help the private sector absorb securities released during balance sheet reduction.

(5) De-stigmatize Fed’s liquidity tools

Removing concerns about using discount windows, standing repo, and intraday overdraft tools can lower banks’ precautionary reserve needs. This requires systematic improvements in transparency, pricing, and communication by the Fed.

Second group: Directly reducing non-reserve liabilities

(1) TGA management recalibration

Reduce the Treasury’s cash buffer in the Fed account from “about 5 days of operations” to “about 2 days,” with the excess transferred back to commercial banks (similar to historical TT&L arrangements). Estimated balance sheet reduction: $200 billion to $400 billion. The paper admits that since deposits would increase correspondingly, the net effect would not be one-to-one.

(2) Lower foreign reverse repo pool attractiveness

Reduce interest payments and set caps to encourage foreign central banks and sovereign funds to shift funds from the Fed’s RRP pool to U.S. Treasuries. Estimated impact: $0 to $100 billion, with effectiveness depending on external cooperation.

Waugh’s Signal: From Technical Paper to Policy Expectation

Understanding this paper requires considering the Fed’s personnel background. Market consensus expects Waugh to succeed as Fed Chair. Waugh has long criticized the Fed’s QE expansion policies and has publicly expressed a preference for balance sheet reduction.

The release of this working paper, led by Milan, is viewed as a forward-looking signal of the future policy orientation during the “Waugh era.” CITIC Securities’ research team notes that given Waugh’s stance and the potential space revealed in this paper, there is a gradual exploration of restarting balance sheet reduction under his leadership.

However, both the paper and Milan’s speech repeatedly emphasize that speed and pace are the most critical operational constraints. Milan explicitly states: “Once reform preparations begin, following the usual pace under the Administrative Procedure Act (APA), it could take over a year, or even several years.” He cites the SLR reform—taking nearly six years from temporary easing to formal regulation—as an example.

This implies that the Fed will not immediately restart balance sheet reduction just because of this paper’s release. The more likely path is to start with less controversial, technically feasible options, while providing market guidance on how new mechanisms might operate.

CITIC Interpretation: Which Are Feasible, Which Are More Ideals

From a practical perspective, CITIC Securities’ team systematically assesses the 15 options and concludes:

Feasible options:

• Relaxation of LCR standards: a technical supervisory reform with controllable variables, giving the Fed reform initiative significant control;
• Reform of standing repo facilities (SRP): directly de-stigmatizes the tool, no legislative changes needed;
• Upgrading Fedwire and payment systems: infrastructure improvements with clear long-term direction;
• ILST supervisory adjustments: some reforms can be implemented through supervisory culture shifts without legal amendments.

More aggressive or externally dependent options:

• Layered reserve interest payments: may trigger nonlinear responses in the banking system, operationally complex;
• TGA management reform: involves coordination between Treasury and Fed, requiring political consensus;
• Lowering attractiveness of foreign reverse repos: highly reliant on external institutions’ willingness, effect uncertain.

Overall, CITIC Securities considers this “a practical and reference-worthy reform menu,” but expects actual implementation to lag behind the potential upper bounds depicted in the paper. It should be viewed as a directional guide rather than an immediate policy commitment.

( Market Impact: Increased Volatility, No Change in Rate Cut Logic

Regarding bond markets, the Fed’s balance sheet reduction effectively reduces base money supply, increasing the scale of U.S. Treasuries that the private sector must absorb. CITIC Securities believes this will amplify market volatility and raise tail risks—though some regulatory easing (like SLR relief) can help expand dealer capacity.

Objectively, the paper advocates against directly selling securities to accelerate reduction; instead, it favors letting maturing securities roll off naturally, while providing dealers and repo markets with higher absorption buffers. This limits short-term shocks.

CITIC Securities judges that U.S. Treasuries are currently more suitable for trading opportunities, with short-term bonds outperforming long-term ones.

For equities, balance sheet reduction exerts contractionary effects via money supply and portfolio effects, but these can be hedged by lowering the federal funds rate. CITIC Securities believes that if reforms proceed, the rate path will need adjustment, but this is only loosely linked to current monetary policy pace. U.S. stocks may wait for correction windows to seek safer margins.

For gold, the impact of balance sheet reduction is unlikely to change the long-term strategic logic of central banks’ gold accumulation, which is driven more by geopolitical reshuffling and dollar reserve diversification. Gold remains a medium- to long-term allocation asset.

Milan explicitly states that the contractionary effects of balance sheet reduction can be hedged through rate cuts, and “may even lead to a larger decline in the federal funds rate relative to baseline scenarios.” CITIC Securities projects U.S. CPI to fluctuate between 3.0% and 3.5% this year, maintaining their view that the Fed will cut rates by 25 basis points in the second half, with no direct link to balance sheet policies.