Quantitative Tightening Around the Globe: What Have We Learned?

Abstract

This paper uses the recent cross-country experience with quantitative tightening (QT) to assess the impact of shrinking central bank balance sheets.We analyze the experience in seven advanced economies (Australia, Canada, Euro area, New Zealand, Sweden, UK and US)documenting different strategies and the substantive reduction in central bank balance sheets that has already occurred.Then we assess the macroeconomic and financial impact of QT announcements on yields and a range of other market prices.QT announcements increase government bond yields, steepening the yield curve and potentially signaling a greater commitment to raising policy interest rates, but have more limited effects on most other financial market indicators.Active QT has a larger impact than passive QT, particularly on longer maturities.The implementation of QT has been associated with a modest rise in overnight funding spreads and a decline in the "convenience yield" of government bonds, but QT transactions did not significantly affect the pricing and market liquidity of government debt securities.Finally, we evaluate who buys assets when central banks unwind balance sheets, an issue which will become increasingly important if central banks continue to reduce their security holdings while government debt issuance remains elevated.We find that increased demand by domestic nonbanks has largely compensated for reduced bond holdings by central banks.This series of cross-country results suggests that QT has had more of an impact than "paint drying", but far less than simply reversing the effects of the quantitative easing programs launched during periods of market stress.Looking ahead, although QT has been smooth to date, frictions could increase in the future so that QT quickly evolves into more like watching "water boil".

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.