Re-Engineering the Yield Curve: How Shifts in Treasury Demand and Supply Shape Rates

 When the U.S. Department of the Treasury issues new debt or the Federal Reserve alters its balance sheet, how much do Treasury yields actually move?

For decades, economists and market participants relied on reduced-form event studies or term structure models to answer this question. However, as illuminated by Federal Reserve Board economists Daniel Beltran and Canlin Li in their research paper, Estimating Yield Impacts of Treasury Demand and Supply Changes, the answer is not static. The true yield impact of a supply or demand shock depends fundamentally on who is on the other side of the trade.

By developing a demand system framework that combines Bayesian time-varying holdings shares with instrumental variable estimations, Beltran and Li uncover a structural transformation in the U.S. Treasury market: the market has become substantially more price-sensitive over time, but this structural shift brings both market-dampening benefits and sudden fragility risks.

The Core Framework: Solving Endogeneity and Time-Varying Shares

Estimating asset demand elasticities in bond markets presents two persistent econometric hurdles:

1. Endogeneity and Simultaneity: Price and quantity are determined simultaneously in equilibrium. Portfolio choices shift yields, and yields concurrently reallocate portfolio choices. Standard OLS regressions yield demand elasticities that are implausibly close to zero.

2. Shifting Investor Base: Most existing literature assumes static holding shares or relies on long-run historical averages. In reality, investor compositions shift dramatically during quantitative easing (QE), regulatory overhauls, and structural global capital flows.

To solve the time-varying holding shares issue, the authors implement a Bayesian Dynamic Linear Model (DLM) across 18 investor sectors. By enforcing exact sum-to-one constraints and modeling latent holding shares as random walks, the DLM smoothly separates Treasury notes and bonds from short-term Treasury bills.

To resolve endogeneity, Beltran and Li employ a Two-Stage Least Squares (2SLS) instrumental variable approach. They instrument the representative 5-year Treasury yield using the bond holding share of price-insensitive investors (the Federal Reserve, Government-Sponsored Enterprises, state/local governments, and official sectors). Because these sectors operate based on policy, regulation, or liquidity mandates rather than return-maximization, their holdings create exogenous shifts in the net residual supply facing private, price-sensitive investors.

The baseline log portfolio demand equation for sector $i$ is specified as:

$$s_{i,t} = \alpha_i + \beta_i \cdot y_t + \lambda_i \cdot s_{i,t-1} + X_t'\delta_i + \epsilon_{i,t}$$

where $s_{i,t} = \ln(H_{i,t}/A_{i,t})$ is the log Treasury bond share of sector $i$'s total financial assets $A_{i,t}$, $y_t$ is the 5-year Treasury yield, $s_{i,t-1}$ captures portfolio rebalancing persistence, and $X_t$ represents sector-specific macro-financial control variables.

The Elasticity Spectrum: Who Moves Market Yields?

The paper reveals immense cross-sectional heterogeneity in yield sensitivity ($\beta_i$) across institutional sectors:

                     ESTIMATED YIELD ELASTICITY SPECTRUM
                     
  Very Price-Sensitive      Price-Sensitive     Somewhat Sensitive      Price-Insensitive
 [Hedge Funds & HHs]      [Broker-Dealers]     [Insurers/Pensions]         [Banks]
      (β = +41)               (β = +12)            (β = +2 to +8)            (β = -6)
         |                        |                      |                     |
  =======+========================+======================+=====================+======

1. Very Price-Sensitive ($\beta_i \approx 41$): Households and Hedge Funds (HHF)—which includes domestic hedge funds, Cayman-domiciled basis trade vehicles, and retail investors—exhibit an elasticity of approximately 41. A 1 percentage point increase in yield induces a 41% expansion in their Treasury portfolio allocation.

2. Price-Sensitive ($\beta_i \approx 10 - 12$): Broker-Dealers ($\beta_i = 12.14$), Exchange-Traded Funds ($\beta_i = 10.37$), ABS Issuers ($\beta_i = 10.16$), and Foreign Private Investors ($\beta_i = 9.60$) demonstrate high responsiveness as market-makers and total-return arbitrageurs.

3. Somewhat Price-Sensitive ($\beta_i \approx 2 - 8$): Property & Casualty Insurers ($\beta_i = 8.27$), Mutual Funds ($\beta_i = 7.29$), Nonfinancial Businesses ($\beta_i = 5.50$), Pension Funds ($\beta_i = 2.69$), Life Insurance ($\beta_i = 2.45$), and Foreign Official Investors ($\beta_i = 2.11$) exhibit modest elasticities, as their demand is tied to duration-matching liabilities, cash management, or foreign exchange reserve policies.

4. Price-Insensitive / Negative ($\beta_i = -6.36$): Banks and Bank Holding Companies display a negative yield elasticity ($\beta_i = -6.36$). Commercial banks hold Treasuries predominantly for liquidity requirements (e.g., LCR compliance) and risk-weighted capital management, continuing to accumulate bonds post-GFC even as market yields fell.

Structural Evolution: Greater Price Sensitivity Meets Fragility Risk

By combining time-varying holding shares with sectoral elasticities, Beltran and Li calculate the aggregate market demand elasticity ($\beta_M = \sum \omega_{i,t} \beta_i$) and its inverse, the market multiplier ($dy / [dS/S] = 1/\beta_M$).

                HISTORICAL SHIFT IN TREASURY MARKET COMPOSITION
                
    2000s Market Structure                   2020s Market Structure
   +-----------------------+                +-----------------------+
   | Foreign Official      |                | Foreign Official      |
   | Investors (FOI)       | (Declined)     | Investors (FOI)       |
   | [Price-Insensitive]   | -------------> | [Price-Insensitive]   |
   +-----------------------+                +-----------------------+
   | Households &          | (Expanded)     | Households &          |
   | Hedge Funds (HHF)     | -------------> | Hedge Funds (HHF)     |
   | [Highly Price-Sens.]  |                | [Highly Price-Sens.]  |
   +-----------------------+                +-----------------------+

Over the past quarter-century, the Treasury market underwent a fundamental structural shift:

• Declining FOI Share: Foreign official institutions, once dominant holders of U.S. debt, saw their market share steadily contract.

• Expanding HHF Share: Highly elastic private players—most notably hedge funds engaging in cash-futures basis trades—expanded their market footprint substantially after 2015 and post-2020.

The Supply Shock Benchmark

Under current market conditions (as of 2026:Q1), a $100 billion increase in Treasury bond supply raises 5-year yields by approximately 3.2 basis points. Normalized as a percentage of public debt, a supply expansion equal to 1% of Treasury bonds outstanding increases 5-year yields by roughly 7.6 to 8 basis points.

The Double-Edged Sword of Market Fragility

In normal times, the presence of highly elastic marginal buyers (like hedge funds) buffers supply shocks; small yield increases prompt quick absorption. However, if these leveraged price-sensitive investors step to the sidelines during market dislocation (such as the March 2020 dash-for-cash), aggregate market elasticity collapses instantly, leaving the market in the hands of inelastic buyers and triggering severe yield volatility.

Policy Simulations: Foreign Sales and Quantitative Tightening (QT)

Beltran and Li apply their estimated framework to evaluate two critical policy scenarios:

Scenario A: $500 Billion Sectoral Asset Sales

The yield impact of asset sales depends critically on who is selling and who remains to clear the market:

• $500 Billion Foreign Official (FOI) Sale: If foreign official central banks dump $500 billion in Treasuries, 5-year yields rise by 13.9 basis points. Because price-sensitive private buyers remain active, they absorb the issuance with modest rate adjustments.

• $500 Billion Household & Hedge Fund (HHF) Sale: If HHF investors liquidate $500 billion and step out of the market, 5-year yields spike by 55.2 basis points. The exit of the market's most elastic buyer leaves only less price-sensitive investors to absorb the debt, requiring a much steeper yield increase to restore equilibrium.

                  YIELD IMPACT OF $500 BILLION SECTORAL SALES
                  
   FOI Sales ($500B)  | 13.9 bps
                      |
   HHF Sales ($500B)  |=========================================== 55.2 bps
                      +-------------------+-------------------+---
                      0                  20                  40 (bps)

Scenario B: Federal Reserve Balance Sheet Runoff

The authors simulate a Quantitative Tightening (QT) scenario where the Fed reduces its Treasury holdings to 5% of GDP over 8 quarters, while gross Treasury debt grows at 5% annually and real GDP expands at 2.5%.

• Cumulative Yield Impact: The Fed's balance sheet runoff contributes 49.5 basis points of upward yield pressure.

• Net Issuance Absorption: Concurrent net fiscal issuance adds 62.5 basis points.

• Total Effect: Combined, the structural shift increases 5-year Treasury yields by 112 basis points over eight quarters.

Mindmap: Key Deductions & Conclusions

The diagram and structured breakdown below outline the core empirical deductions, methodological innovations, and ultimate policy conclusions of the research paper.

Code snippet

mindmap
  root((Fed Treasury Demand & Supply Paper IFDP 1447))
    Methodological Innovations
      Dynamic Linear Model DLM
        Smooth time-varying share estimation
        Enforces sum-to-one constraint
        Splits T-Bonds from T-Bills
      Instrumental Variables 2SLS
        Exogenous instrument: Fed + Inelastic Holders Share
        Controls for macro-financial confounders
    Sectoral Elasticity Deductions
      Very Price-Sensitive
        Households & Hedge Funds β = 41
      Price-Sensitive
        Broker-Dealers β = 12
        ETFs, ABS, Foreign Private β = 10
      Somewhat Sensitive
        Property-Casualty β = 8
        Mutual Funds β = 7
        Foreign Official β = 2
      Price-Insensitive
        Commercial Banks β = -6
    Market Structure Evolution
      Declining Foreign Official Share
      Expanding Hedge Fund Basis Trade Share
      Aggregate Market Sensitivity Increased
    Policy Counterfactual Conclusions
      100 Billion Supply Shock
        +3.2 bps 5-Year Yield Increase
      500 Billion Asset Sales
        Foreign Official Sale: +13.9 bps
        Hedge Fund Sale: +55.2 bps
      Fed QT Balance Sheet Runoff
        Runoff Effect: +49.5 bps
        Net Issuance Effect: +62.5 bps
        Total Yield Impact: +112 bps

Key Deductions

• Heterogeneous Demand Elasticity Is Essential: Treating Treasury market participants as a homogeneous group introduces severe aggregation bias. Highly elastic hedge funds ($\beta_i \approx 41$) operate under entirely different return-risk and leverage frameworks than price-insensitive commercial banks ($\beta_i \approx -6$) or reserve-managing foreign central banks ($\beta_i \approx 2$).

• Market Multipliers Are Dynamic: The slope of the aggregate Treasury demand curve is governed by time-varying holding weights ($\omega_{i,t}$). As price-sensitive entities capture larger market shares, the baseline market multiplier shrinks, dampening everyday yield sensitivity to standard supply shocks.

• Validation via Explanatory Power: Regression models incorporating changes in the market multiplier and surprise investor purchases account for over 71% ($R^2 = 0.714$) of quarterly variation in 5-year Treasury yields from 2000 to 2026, confirming that investor composition directly shapes rate trajectories.

Primary Policy Conclusions

• Quantitative Tightening Imposes Quantifiable Costs: Private market absorption of Fed balance sheet unwind is achievable, but requires higher equilibrium interest rates (approximately +112 bps over 2 years under a 5% GDP target).

• Foreign Official Liquidation Is Manageable: Dumps of Treasury reserves by foreign central banks yield moderate, non-disruptive yield increases (+13.9 bps per $500B) provided domestic total-return arbitrageurs remain capitalized and willing to buy.

• Structural Fragility Requires Regulatory Vigilance: Because the Treasury market relies increasingly on highly elastic leveraged investors (hedge funds), financial stability depends heavily on maintaining balance sheet intermediation capacity. If liquidity or funding shocks force hedge funds to exit, Treasury yields become vulnerable to severe price dislocation.