开放式国债购买：从市场运作到金融宽松（英）

Stefania D’Amico|Max Gillet|Sam Schulhofer-Wohl|Tim Seida Open-Ended Treasury Purchases: From Market Functioning to Financial EasingStefania D’Amico, Max Gillet, Sam Schulhofer-Wohl, and Tim SeidaFederal Reserve Bank of New York Staff Reports, no.1183February 2026https://doi.org/10.59576/sr.1183 Abstract We assess whether the Fed’s asset purchases can be tailored to either restore market functioning orprovide economic stimulus. When the communicated goal is restoring market functioning and purchases’implementation is flexible, flow effects are significant: relative price deviations narrow. However, stockeffects remain near zero and hence not stimulative. When the communicated goal links purchases to theachievement of the dual mandate, improving their size’s predictability, stock effects rise consistentlyabove zero. When the communicated implementation improves the predictability of the purchases’maturity composition, stock effects become large. Jointly, the communicated goal and implementationcan shape the purchases’ effects. JEL classification:E43, E44, E52, E58Keywords:quantitative easing, market-functioning asset purchases, communicated policy goals, assetpurchases implementation This paper presents preliminary findings and is being distributed to economists and other interestedreaders solely to stimulate discussion and elicit comments. The views expressed in this paper are those ofthe author(s) and do not necessarily reflect theposition of the Federal Reserve Bank ofChicago, theFederal Reserve Bank of Dallas, the Federal Reserve BankofNew York or the Federal Reserve System.Any errors or omissions are the responsibility of the author(s). To view the authors’ disclosure statements, visithttps://www.newyorkfed.org/research/staff_reports/sr1183.html. 1Introduction Central banks have employed large-scale asset purchases (LSAPs) for two distinct objectives:to provide monetary policy accommodation when the policy rate is near zero, and to supportmarket functioning or financial stability.1 A fundamental unresolved question in the litera-ture is how LSAPs can be tailored to accomplish different objectives, or, put differently, howthe same purchase operation can be used to achieve different goals in different economic andmarket conditions. This question has been highly debated in policy and academic circles.2 The answer depends on two crucial factors: (1) whether the LSAP’s effects are affected byits communicated objective and (2) how the LSAP’s implementation is adjusted to achievethat objective.Our focus is to analyze these two key factors using the ideal setting pro-vided by the$2.9 trillion of Treasury purchases conducted by the Federal Reserve (the Fed)in 2020-2022, which served multiple objectives over time.The initial goal was to supportsmooth market functioning, but later the goal was expanded to foster accommodative finan-cial conditions as in previous QE.3As a result, the purchases’ implementation was greatlychanged. The purchase pace, which initially peaked around$75 billion per day, was reducedto$80 billion per month once the goal shifted. To understand whether the program’s communicated goal and implementation matteredfor the achievement of each objective, we estimate how the purchases’ stock and flow effectsevolved with the communicated goal and communicated implementation.4 To this end,we use an approach similar to D’Amico and King (2013), which we modify to account forthe several novel features of the 2020-2022 purchase program.This program was open-ended rather than fixed in size.The purchase pace was characterized by unprecedentedscale and speed. During March 2020, the New York Fed’s Open Market Trading Desk (the Desk) bought in four days the same amount purchased during the seven-month-long 2009Treasury LSAP. Finally, commitment and flexibility were key aspects of the program. TheFed committed to purchase securities “in the amount needed,” in effect putting no limit. Thepurchases’ size and maturity distribution were adjusted weekly and even daily as needed.This, at least initially, could have reduced the predictability of the total size and compositionof purchases. In open-ended programs, investors’ ability to predict the total reduction in privately-heldsecurities and duration risk should be key to the stock effect—the permanent price impactof expected changes in the stock of debt and duration risk.How broadly these expectedchanges affect the yield curve is determined by the cross-price sensitivity of each securityto the purchases of securities with similar maturity—the substitution effects—which in theaggregate move down entire sectors of the yield curve, providing monetary stimulus (D’Amicoand King (2013)). In principle, substitution effects should be weak when markets are highlysegmented. Hence, stock effects should grow larger as investors’ ability to predict the totalsize and composition of purchases improves and market segmentation decreases. In contrast, the flow effect mea