The demand for government debt
(June 2023, revised December 2025)
Summary
Focus
We investigate how central bank balance sheet policies affect investor demand for government debt. As central banks embark on quantitative tightening (QT) to shrink their balance sheets, how different sectors absorb government debt will largely depend on how sensitive their demand is to changes in the yields (or prices) of government bonds. We study compositional shifts in the investors that hold government debt in the euro area, Japan, the United Kingdom and the United States, and consider how different investor groups would respond to changes the supply of government bonds.
Contribution
We estimate the yield demand elasticity for government bonds across different sectors and how this shapes the way that the government bond market adjusts when central banks implement balance sheet policies. To set the stage, we first run regressions that measure the marginal absorption by different investor groups as the total amount of government debt outstanding changes. We then draw on a demand system approach to compute the yield sensitivity. To account for any endogeneity between latent demand by various investor groups and government bond yields, we rely on monetary policy surprises as instruments for the yields in two-stage least squares regressions.
Findings
We estimate that a 1% increase in the central bank holdings of US Treasuries results in a drop in long-term yields of around 8–13 basis points, depending on the market composition. While the elasticity of individual sectors remains consistent regardless of whether the central bank's share in the Treasury market is increasing or decreasing, quantitative easing (QE) and QT programs can have asymmetric effects due to differences in market composition. The impact of QE on yields is estimated to be larger, as overall demand elasticity in the Treasury market was lower during QE periods.
Abstract
We document that the sectoral composition and marginal buyers of government debt differ notably across jurisdictions and over time. We use instrumental variables derived from monetary policy surprises to estimate the demand elasticities of various sectors. In the United States, commercial banks and mutual funds exhibit the most price-elastic demand, while the foreign official sector has a price-inelastic demand. Based on these estimates and under certain assumptions, we find that a 1% increase in the central bank holdings of US Treasuries results in around 8 to 13 basis point drop in long-term yields depending on the market composition. Elasticities of individual sectors do not differ in a statistically significant manner when the central bank share in the Treasury market increases or decreases. However, different market compositions during various quantitative easing (QE) and quantitative tightening (QT) programs have led to an asymmetric effect with the impact of QE on yields being greater than that of QT. Our results suggest that, overall, the demand for US Treasuries is considerably more elastic than for equities, corporate bonds and emerging market sovereign bonds found in the literature. We also repeat the analysis for other jurisdictions and compare estimates for different sectors.
JEL classification: E58, G11, G21, G23, H63
Keywords: government debt, demand, price elasticity, quantitative easing, quantitative tightening
