Prolepsia forecastYes / no
Will the Federal Reserve's balance sheet be larger on 31 December 2027 than on 31 December 2026?
As of 16 November 2026, Prolepsia puts the chance of yes at 68%.
68%chance of yes
How Prolepsia reasons
The report Prolepsia wrote with this forecast.
How forecasts are made · Track record
TL;DR
The forecast is 67.6% for Yes and 32.4% for No. Current policy favors maintaining securities principal and adding assets when needed to supply ample reserves, but the purchase pause and forthcoming policy review leave a substantial risk of flat or declining total assets.
Key evidence
- The operative policy supports maintenance or growth. The October 28 directive requires full rollover of Treasury principal, reinvestment of agency principal into Treasury bills, and additional purchases when appropriate to maintain ample reserves. Renewed ordinary principal runoff would require a policy change (implementation note).
- Permission to buy is not a commitment to expand. Four successive operation periods spanning August 14–December 11 contain no reserve-management purchases. The approximately $14 billion of reinvestments in the latest period replaces principal rather than representing equivalent net growth (purchase schedules).
- A reason for purchases to resume exists. Treasury projects cash balances rising from $850 billion at the end of December 2026 to $1,000 billion at the end of March 2027. That would drain $150 billion of reserves, other things equal, although it does not require a matching asset purchase (Treasury estimates).
- Policy change faces a calendar constraint. Chair Warsh anticipated balance-sheet recommendations early in the first quarter of 2027, not immediate implementation. Governor Miran warned that preparatory reforms could take well over a year after a decision (press conference; Miran speech).
Uncertainties
The decisive unknown is whether the review changes actual reinvestment policy soon enough to reduce assets by the resolving date. Smaller operational changes could arrive faster than a comprehensive redesign.
A continued purchase pause could also produce a No result through accounting changes or lower lending, even without renewed quantitative tightening. Conversely, temporary lending can increase total assets without quantitative easing. The comparison uses consolidated total assets on December 29, 2027, against December 30, 2026—not today’s balance sheet, reserves, or the securities portfolio alone. Both future snapshots remain unknown.
Read the full report
TL;DR
The forecast is 67.6% for Yes and 32.4% for No. The central case is a larger Federal Reserve balance sheet at the specified end-2027 observation, supported by principal preservation and purchases to meet reserve needs. The key distinction is total assets, not quantitative easing: reserve-management purchases and lending count, while replacement purchases alone do not establish growth.
Context
This forecast, dated November 16, 2026, compares consolidated total assets on December 29, 2027, with December 30, 2026. These are the final Wednesday observations of the two years. Any increase qualifies; equality resolves No. The measure is the Wednesday level in the Federal Reserve’s consolidated statement, not a weekly average, bank reserves, or securities holdings alone. Official revisions available by January 31, 2028, apply to the comparison.
The latest reported total is $6,758,867 million for November 11, published in the November 12 release (H.4.1). That establishes the current scale, not the resolving baseline. The baseline remains in the future. Current policy preserves securities principal, but additional purchases have paused. The question is whether reserve needs bring purchases back before policy changes or other asset movements produce a lower year-end total.
Evidence
History provides a modest starting point. Across 22 completed annual comparisons covering 2004–2025, consolidated assets increased in 12, or 54.5% of cases. This is a broad reference class, not a reliable stand-alone forecast: it combines crisis expansion, quantitative easing, maintenance periods, and deliberate contraction (WALCL historical series). The current operating instructions and the timing of possible changes carry more weight than that mixed historical record.
Those instructions support maintenance or growth. The October 28 directive requires full rollover of Treasury principal and reinvestment of agency principal into Treasury bills. It also permits additional purchases when appropriate to maintain ample reserves. Ordinary principal runoff is therefore not the current instruction (implementation note). This is the strongest reason to favor Yes: an unchanged framework preserves principal while allowing assets to grow as demand for Federal Reserve liabilities increases.
Actual operations put a clear limit on that argument. Additional reserve-management purchases have fallen to zero. The complete supplied purchase-plan history is below. Amounts are billions of dollars, and the rows are successive operation periods spanning December 12, 2025–December 11, 2026. These are scheduled amounts, not realized calendar-month totals (New York Fed operational details).
| Successive operation period | Additional reserve-management purchases | Principal reinvestments |
|---|---|---|
| 1 | 40 | 14.4 |
| 2 | 40 | 15.4 |
| 3 | 40 | 13.4 |
| 4 | 40 | 13.8 |
| 5 | 25 | 15.5 |
| 6 | 10 | 16.3 |
| 7 | 10 | 16.5 |
| 8 | 10 | 17.6 |
| 9 | 0 | 17.0 |
| 10 | 0 | 15.6 |
| 11 | 0 | 14.6 |
| 12 | 0 | 14.0 |
The final four periods span August 14–December 11. Their zero additional purchases show why early expansion cannot simply be extrapolated into the forecast year. The latest $14.0 billion of reinvestments replaces agency principal; it is not $14.0 billion of net growth (purchase schedules). Preserving principal is not the same as increasing total assets.
The reason for the pause matters. In September, SOMA manager Roberto Perli said assessed reserve demand had changed little. Lower projected Treasury cash balances had instead increased expected reserve supply. He also reported that roughly $400 billion of preceding net bill issuance had generated only modest funding pressure (Perli speech). I read this as an operational pause under the existing framework, not proof that structural reserve demand has fallen enough to support sustained contraction.
Treasury cash management supplies a concrete reason for conditions to change. Its November financing estimates assumed $850 billion of cash at the end of December 2026 and $1,000 billion at the end of March 2027. That implies a prospective $150 billion reserve drain, other things equal (Treasury estimates). It creates a reason to resume purchases after the baseline date. It does not imply a dollar-for-dollar response, establish December 2027 cash needs, or guarantee that purchases made for seasonal reasons remain outstanding at year-end.
Longer-run official projections also support growth, but only conditionally. The annual open-market-operations report, using January 2026 assumptions, illustrated reserve-management purchases starting at $345 billion annually and ranging from $330–395 billion. The subsequent pause shows why these figures are a framework illustration rather than a current commitment (annual report). They explain the mechanism behind eventual expansion; they do not settle this particular year-end comparison.
Bank evidence gives little reason to assume reserve demand will collapse automatically. The March survey covered 92 respondents holding about three-fourths of reserves. Most reported no change in preferred reserves after removal of the standing-repo aggregate cap (Senior Financial Officer Survey). Easier access to contingent funding can reduce the need to hold reserves, but access and actual willingness to rely on it are different things. Survey responses are also institution counts, not a dollar estimate of releasable reserves.
The strongest downside is the policy review. At the October 28 press conference, Chair Warsh anticipated recommendations early in the first quarter of 2027. Committee consideration would follow; the statement did not adopt a contraction target or start date (press-conference transcript). A decision to resume runoff or otherwise lower desired reserves could change the direction of assets during the forecast horizon.
Implementation time limits that risk. Governor Miran warned that preparatory reforms could take well over a year after a decision, possibly several years (Miran speech). Research identifies substantial eventual reduction capacity, but that capacity depends on reforms rather than representing an adopted near-term program (Federal Reserve working paper). Governor Barr has also argued against treating shrinkage as an objective in itself, citing liquidity and payment-system trade-offs (Barr speech). These are constraints, not guarantees against contraction.
Finally, accounting and lending can decide a close result. The latest release shows that year-over-year movements in securities premiums and discounts together reduced assets by $24.632 billion. Full principal reinvestment therefore does not hold total assets perfectly constant (H.4.1). Temporary lending works in the opposite direction. The Fed reported $74.6 billion of repos on December 31, 2025, which fell to zero on January 7, 2026 (historical H.4.1 release). A reporting-date lending spike can materially change the answer without changing the underlying securities policy.
What's non-obvious
The obvious debate is whether the Fed will expand or shrink its securities portfolio. The resolving question is broader. Replacing securities with repo or discount-window assets need not reduce consolidated total assets. Likewise, ending runoff does not guarantee growth when purchases remain paused and accounting values decline. The decisive issue is the combined movement of purchases, principal repayments, lending, and other assets between two specific observations.
Timing is the second less obvious edge. A proposal for a much smaller eventual balance sheet is not the same as contraction realized during this comparison year. Preparatory work can delay reductions, while reserve demand can require purchases in the meantime. The reverse also holds: a temporary expansion during 2027 does not ensure Yes if it unwinds before the final Wednesday. The forecast favors growth without requiring a new quantitative-easing program or dismissing the policy-review risk.
Uncertainties
The largest gap is the content and implementation schedule of the balance-sheet review. No verified adopted 2027 runoff target is established in this analysis. Final recommendations, subsequent operating instructions, and effective dates for any liquidity reforms would distinguish an eventual shrinkage ambition from a program that changes the resolving total.
Reserve demand is also uncertain. Fresh autumn survey results and an exact current official end-2027 projection were not established. The available survey is informative but dated. A complete inventory of enacted liquidity reforms was not verified, so that gap is not evidence that no further reform exists. Updated bank responses and sustained funding-market behavior would clarify how long purchases can remain paused.
Both endpoints remain exposed to Treasury cash movements, temporary lending, settlement timing, accounting adjustments, and revisions. Stress that enlarges the December 2026 baseline and then reverses can favor No; lending outstanding at the December 2027 observation can favor Yes. These risks preserve a meaningful downside even with principal-preserving policy. They do not change the saved forecast: 67.6% for Yes and 32.4% for No.
Forecast history
| Date | Forecast |
|---|---|
| 68% chance of yes |
Sources Prolepsia read
- Federal Reserve Balance Sheet: Factors Affecting Reserve Balances - H.4.1 - November 12, 2026www.federalreserve.gov
- Assets: Total Assets: Total Assets (Less Eliminations from Consolidation): Wednesday Level (WALCL) | FRED | St. Louis Fedfred.stlouisfed.org
- Treasury Securities Operational Details - FEDERAL RESERVE BANK of NEW YORKwww.newyorkfed.org
- Treasury Announces Marketable Borrowing Estimates | U.S. Department of the Treasuryhome.treasury.gov
- Modifications to the Enhanced Supplementary Leverage Ratio Standards for U.S. Global Systemically Important Bank Holding Companies and Their Subsidiary Depository Institutions: Final Rule | OCCwww.comptrollerofthecurrency.gov
- Discount Window Modernizationwww.frbdiscountwindow.org
- The Federal Reserve’s Balance Sheet Size and “Other Deposits”: Some Financial Stability Considerations - Federal Reserve Bank of Chicagowww.chicagofed.org
The question
This question asks whether the total assets held by the United States Federal Reserve System will be greater at the end of 2027 than at the end of 2026, measured in nominal U.S. dollars. The forecast is made as of November 16, 2026, and compares the Federal Reserve's consolidated balance sheet at two year-end reference dates: December 31, 2026, and December 31, 2027. The relevant measure is the total assets of the Federal Reserve Banks, as reported in the Federal Reserve's weekly H.4.1 statistical release, 'Factors Affecting Reserve Balances of Depository Institutions and Condition Statement of Federal Reserve Banks.' This includes Treasury securities, agency mortgage-backed securities, loans, and other assets, rather than exclusively securities held for monetary policy purposes. As background, the Federal Reserve's balance sheet expanded dramatically during the COVID-19 pandemic, approaching $9 trillion in 2022, before subsequently contracting through quantitative tightening. By November 2026, its total assets were approximately $6.76 trillion, according to the Federal Reserve's November 12, 2026, H.4.1 release. The future size of the balance sheet depends on Federal Reserve decisions regarding securities holdings, reinvestments, reserve management, lending facilities, and broader financial conditions. The question concerns the actual change in total reported assets between the two year-end reference dates, regardless of the monetary policy decisions or economic developments responsible for that change. Because the Federal Reserve reports weekly Wednesday balance sheet observations rather than daily figures for every calendar date, the question will use the final Wednesday observation of each year: December 30, 2026, and December 29, 2027. These provide consistent, publicly verifiable year-end measurements. Source: Federal Reserve H.4.1 statistical release (https://www.federalreserve.gov/releases/h41/).
How it resolves
The question resolves YES if the Federal Reserve System's consolidated total assets, as reported for December 29, 2027, are strictly greater than its consolidated total assets reported for December 30, 2026. It resolves NO if the December 29, 2027, value is less than or equal to the December 30, 2026, value. The primary resolution source will be the Federal Reserve's official H.4.1 statistical releases, specifically the 'Total assets' entry in the table titled 'Consolidated Statement of Condition of All Federal Reserve Banks,' using the Wednesday observation rather than the weekly average. The December 30, 2026, observation is expected to appear in the December 31, 2026, release, and the December 29, 2027, observation is expected to appear in the December 30, 2027, release, subject to the Federal Reserve's publication schedule. Values will be compared in their published units, ordinarily millions of U.S. dollars, without inflation adjustment. If the Federal Reserve revises either observation, the latest officially revised figures available by January 31, 2028, will be used. If the relevant releases are unavailable or their publication is delayed, official Federal Reserve historical H.4.1 data or the corresponding Federal Reserve Economic Data (FRED) series WALCL, 'Assets: Total Assets: Total Assets (Less Eliminations from Consolidation): Wednesday Level,' may be used as a fallback, provided the observations are equivalent to the specified H.4.1 measure. The Federal Reserve's original data take precedence over third-party reproductions if discrepancies arise. If the Federal Reserve changes the presentation or classification of its balance sheet during the forecast period, the most directly comparable official consolidated total-assets series will be used, with any officially published historical restatements applied consistently to both dates. If no sufficiently comparable official data are available by March 31, 2028, the question will be annulled rather than resolved based on estimates or assumptions. The outcome is determined exclusively by the comparison of the two reported asset totals; no minimum percentage change or dollar increase is required for a YES resolution.
Fine print
The term 'balance sheet larger' refers exclusively to the asset side of the Federal Reserve System's consolidated balance sheet, not its securities portfolio alone, the monetary base, bank reserves, or the combined assets of the Federal Reserve and the U.S. Treasury. The comparison uses nominal U.S. dollars and does not account for inflation or changes in GDP. Temporary increases or decreases in Federal Reserve assets count if they are reflected in the specified year-end observations, even if they subsequently reverse. The question does not require the Federal Reserve to announce quantitative easing, end quantitative tightening, or make any particular policy decision. December 31, 2027, falls on a Friday, so the December 29, 2027, Wednesday observation is used as the closest regularly reported Wednesday balance sheet snapshot. December 30, 2026, is the corresponding final Wednesday of 2026. A change of zero, including equality at the precision published by the Federal Reserve, resolves NO. If extraordinary circumstances cause the Federal Reserve to cease operating in its existing institutional form, an officially reported successor balance sheet may be used only if it is directly comparable across both reference dates; otherwise, the question will be annulled.
More public forecasts
Will the Federal Reserve raise the federal funds target range at any FOMC meeting between 1 January and 31 December 2027?
82%chance of yes
Prolepsia forecast
Will the Federal Reserve cut its target rate at least twice during 2027?
30%chance of yes
Prolepsia forecast
What will the US 10-year Treasury yield be on 31 December 2027, as published by the Federal Reserve (FRED series DGS10)?
5.01median
Prolepsia forecast
Will the World Bank Pink Sheet urea price for December 2027 be higher than its price for December 2026?
45%chance of yes
Prolepsia forecast
Will the Reserve Bank of Australia's cash rate be below 3.5% on 31 December 2027?
13%chance of yes
Prolepsia forecast
Will a US federal government shutdown begin between 12 December and 31 December 2026?
29%chance of yes
Prolepsia forecast