Recently, I’ve been paying close attention to a policy trend from the central bank that warrants in-depth understanding. Federal Reserve Board member Milan and his team recently released a working paper that systematically outlines possible paths for the Fed’s balance sheet reduction. The significance of this paper goes beyond technical details; it signals a major shift in future policy direction.

Market observers on Wall Street should note that the timing of this paper’s release is quite sensitive. The U.S. Senate Banking Committee is about to hold hearings on Kevin Wirth’s nomination as Fed Chair. Wirth has long been critical of the Fed’s large balance sheet, and this paper is widely seen as a forward-looking signal of the policy orientation in the “Wirth era” to come.

There has been a deeply ingrained market perception: the ceiling for Fed balance sheet reduction is the depletion of reserve balances. But this paper breaks that logic. The key insight is that reserve demand itself can be shaped by policy. In other words, the Fed can fully adjust regulatory frameworks and operational mechanisms to achieve significant balance sheet shrinkage while maintaining ample reserves.

The core diagnosis of the paper is particularly interesting. It shifts the perspective from the “supply side” to the “demand side,” pointing out that reserve demand is not an exogenous constraint but is artificially elevated by regulatory rules, supervisory enforcement, and the Fed’s operational framework. Specifically, three mechanisms drive up reserve demand: first, interest rate differentials make reserves a “risk-free earning asset”; second, multiple liquidity management tools create a “ratcheting effect,” with various rules intertwined; third, the discount window has been long stigmatized, leading banks to hoard reserves rather than use them.

So, how much can it be reduced? The paper performs a quantitative estimate based on the Fed’s balance sheet data as of March 11, 2026. At that time, total assets were about $6.646 trillion, with reserves around $3.073 trillion. Using a Monte Carlo aggregation of 15 policy options, the paper estimates reserve demand could decrease by $825 billion to $1.75 trillion, and total assets could shrink by $1.15 trillion to $2.125 trillion. In terms of GDP ratio, the current Fed balance sheet is about 21%, and the estimate suggests it could fall back to levels close to 2012 or 2019, roughly 15% to 18% of GDP.

How to implement specifically? The paper divides the 15 tools into two main categories. The first aims to lower the equilibrium reserve demand, including relaxing LCR standards, reforming standing repo facilities, upgrading Fedwire, and adjusting supervisory criteria. The second directly targets reducing non-reserve liabilities, such as adjusting the Treasury’s cash buffer in the Fed’s accounts and lowering the attractiveness of foreign reverse repo pools.

CITIC Securities’ research team has evaluated the feasibility of these options. They believe that relaxing LCR standards, reforming standing repo facilities, upgrading payment systems, and adjusting ILST supervisory criteria are relatively feasible. However, tiered interest on reserves, TGA management reforms, and reducing foreign reverse repo pools require more external cooperation and are more challenging to implement.

It’s worth noting that both the paper and Milan’s speeches repeatedly emphasize the importance of pace and rhythm. According to the usual procedures under the Administrative Procedure Act, once reserve reforms are initiated, they could take over a year or even several years to implement. For example, the SLR reform took nearly six years from temporary easing to formal regulation. This suggests that the Fed is unlikely to immediately restart balance sheet reduction solely based on this paper; instead, it will likely start with less controversial, technically feasible options.

What about market impacts? Balance sheet reduction essentially reduces base money supply, which will inevitably increase the scale of U.S. Treasury securities that the private sector needs to absorb, amplifying market volatility. However, the paper explicitly opposes accelerating reduction through outright securities sales; a more feasible approach is allowing maturing securities to roll off naturally. From the bond market perspective, U.S. Treasuries are currently more suitable for trading opportunities, with short-term bonds possibly better than long-term bonds.

For the stock market, balance sheet reduction exerts contractionary effects through two pathways: the impact on money supply and portfolio rebalancing. However, these can be offset by lowering the federal funds rate. Milan explicitly states that the contractionary effects of balance sheet reduction can be hedged through rate cuts, and “balance sheet reduction may lead to a larger decline in the federal funds rate relative to the baseline scenario.”

Regarding gold markets, reforming balance sheet reduction is unlikely to fundamentally change the global central banks’ strategic accumulation of gold. The drivers for gold are more related to geopolitical reshuffling and the diversification of dollar reserves, with medium- to long-term allocation value still intact.

Overall, this is a very pragmatic reform menu, but the actual implementation pace will likely be much slower than the potential upper bounds described in the paper. It’s expected that U.S. CPI will fluctuate between 3.0% and 3.5% within the year, with no direct link between balance sheet reduction and rate cuts. The most important significance of this paper is that it provides the market with a new policy imagination space and hints at future Federal Reserve policy exploration directions.