核心问题
Is the inflation target anchoring expectations, or is the central bank merely announcing a number after the private sector has moved on?
Inflation targeting is a number and a public contract about how policy responds when prices move away from the target.
The hard question is speed. Returning too slowly risks credibility; returning too quickly can turn a price shock into unnecessary unemployment.
政策目标
Is the inflation target anchoring expectations, or is the central bank merely announcing a number after the private sector has moved on?
Inflation targeting anchors price-setting through a public target, a reaction function, and the credibility to return misses to target without crushing the economy.
The target works only if households, firms, markets, and wage setters believe misses will be temporary.
工具与手段
政策真正可以调动的杠杆。每种工具都有各自的操作目标、信息需求和政治成本。
Published target
The target states the nominal anchor and gives private plans a reference point.
Policy-rate reaction
Rates move to make the target credible when inflation pressure is persistent.
Forecast and communication
Projections, minutes, speeches, and press conferences explain the expected path back.
Framework review
Periodic reviews reset how the target is interpreted when the economy changes.
传导渠道
政策工具影响价格、产出、就业、信贷、对外收支或产能的渠道。
工资和价格设定中的锚
通胀目标主要通过人们写进合同的数字发挥作用。如果企业定价、工人谈判工资时都假定几年后通胀约为2%,暂时性冲击就不会演变为持久冲击,央行也只需较小的利率调整就能让通胀回归。
实际利率与支出
当通胀超过目标时,央行会把政策利率提高到足以推高实际利率、抑制支出的程度。可信的目标让这一过程代价更低:如果预期通胀保持不变,名义利率每提高一次,实际利率的升幅就更大,因此较小的紧缩就能使需求放缓同样的幅度。
长期利率
长期利率包含对贷款期限内预期通胀的补偿。公众相信的目标能让这部分补偿保持低位且稳定,从而降低住房抵押贷款和企业借款的成本。可信度一旦丧失,长期收益率最先反映出来:在美国,1981年全年通胀在下降,10年期利率却持续上升。
时滞
效果究竟何时显现。有些价格会立即调整;数量、合同、招聘和投资通常反应得更晚。
Expectation reset
Days to months. Communication can move markets quickly, but household and firm expectations are stickier.
Rate transmission
Quarters. Demand-sensitive sectors react first; wage and services inflation often react later.
Credibility repair
Years. Once the target loses credibility, a clean return may require a costly slowdown.
制度约束
决定一项技术上自洽的政策能否实际使用的法律、政治、资产负债表和操作层面的限制。
Target definition
Headline, core, range, point target, average target, and horizon all change the reaction function.
Mandate balance
Employment, financial stability, and distributional stress can change the acceptable speed of return.
Communication discipline
A target without clear explanations of misses becomes a slogan rather than an anchor.
失灵方式
政策可能失灵、来得太晚或作用于错误边际的各种方式。
Targeting through a supply shock
A mechanical return path can damage output when the price shock is sectoral and temporary.
Looking through too much
Repeated misses can teach the private sector that the target is optional.
Communication drift
If every miss gets a new explanation, the framework stops disciplining expectations.
取舍
任何有用的干预都会牵动别的东西。
以多快的速度让通胀回归
目标说明通胀应处于什么水平,却没有规定必须以多快的速度回到那里。供给冲击后急于回归,可能为一次本会自行消退的价格波动付出产出和就业的代价;回归太慢,又可能让人们认定偏离会持续。央行选择的时间跨度以及它如何解释这一选择,正是灵活通胀目标制的实质。
点目标、区间,还是平均
点目标便于沟通。区间目标,例如加拿大的1%至3%,承认通胀会波动。平均目标则承诺弥补过去的偏离。美联储在2020年采用了平均目标,在随后的通胀使这一承诺难以落实之后,于2025年取消了弥补要素。
反通胀的代价由谁承担
通过抑制需求来压低通胀,会让一些人失去工作和工时,而所有人都能从更低的通胀中受益。对利率敏感行业的工人以及与劳动力市场联系薄弱的人,往往承担更多代价,这也是回归速度之所以需要争论、而不只是计算的部分原因。
需要监测的数据
关闭本页之后,继续追踪那些可能推翻这一诊断的数据序列。
- Core inflationCore inflationCore measures help separate persistent pressure from volatile relative-price moves.
- Inflation expectationsInflation expectationsTargets are about expectations as much as realized prices.
- Wage pressureWage growthWages help distinguish one-off price shocks from broad nominal pressure.
历史案例
检验这项政策诊断的真实案例。
1990
New Zealand adopts inflation targeting
A small open economy made the explicit target a central-bank operating regime.
2012
Fed formalizes 2 percent PCE inflation goal
The target became explicit in the Fed's longer-run goals statement.
2020
Flexible average inflation targeting
The Fed changed the framework just before the pandemic inflation cycle tested it.
2025
Fed returns to flexible inflation targeting
The August 2025 revision of the Fed's framework statement dropped the 2020 makeup strategy and the employment "shortfalls" language.
The target is a reaction function, not a slogan
A credible inflation target tells the public how the central bank reacts to misses. It needs a forecast horizon, a measure of inflation, a tolerance for temporary shocks, and a language for explaining why the return path is fast or slow. In the United States these elements are set out in the FOMC's Statement on Longer-Run Goals and Monetary Policy Strategy, first adopted in 2012 and revised in 2020 and again on August 22, 2025 [1]. The current statement names the goal, 2 percent inflation in the price index for personal consumption expenditures; commits to a balanced approach when the employment and inflation goals conflict; and promises a public review of strategy, tools, and communications roughly every five years [2].
The target is doing work only when price setters and wage bargainers believe a miss will not persist. If the public hears a number but cannot infer the reaction function, the anchor is weaker than it looks. New Zealand built the first formal regime of this kind. The Reserve Bank of New Zealand Act came into effect in February 1990; the first Policy Targets Agreement between the Minister of Finance and the Governor set a target band of 0 to 2 percent, and the law made clear that the Governor could lose the job for inadequate performance against it [3].
Supply shocks test the horizon
A relative-price shock creates a dilemma. Tightening enough to force headline inflation back immediately can sacrifice output for a shock that monetary policy cannot undo. Looking through the shock can be right if expectations remain anchored. A target range gives some room for that judgment: Canada's framework, renewed for January 2027 to December 2031, keeps a 2 percent target at the midpoint of a 1 to 3 percent control range [4].
The danger is repetition. If one supply shock follows another, firms and workers may stop treating misses as temporary, and the central bank then has to tighten against the wage-price process, not against the original oil, food, or import-price move. The U.S. surge that began in 2021 shows both halves. Blanchard and Bernanke found that most of it came from shocks to prices, such as sharp increases in commodity prices and sectoral price spikes, not from overheated labor markets, but that the effects of tight labor markets on wages and inflation last longer than those of product-market shocks [5]. The Fed itself expected through the end of 2021 that inflation would subside fairly quickly without a sharp tightening; when that proved wrong it raised its policy rate by 5.25 percentage points over 16 months [6].
Average targeting, 2020-2025
Average inflation targeting promises that past misses matter for future policy. In August 2020 the FOMC said it would seek inflation that averages 2 percent over time: after periods when inflation had been running persistently below 2 percent, policy would likely aim for inflation moderately above 2 percent for some time. It also replaced "deviations" from maximum employment with "shortfalls" [7]. The design answered the problem of the 2010s, when the policy rate spent seven years at its effective lower bound after the global financial crisis.
The world it was built for did not last. The post-pandemic reopening brought the highest inflation in 40 years. In the revision of August 22, 2025, the Committee returned to flexible inflation targeting, eliminated the makeup strategy, and removed "shortfalls" from its statement. Chair Powell said that the idea of an intentional, moderate overshoot had proved irrelevant, since nothing about the inflation that followed was intentional or moderate [8]. A makeup promise is easiest to keep when inflation is low; it is hardest to explain when inflation is already high.
情景 · 解读政策的方法
在套用政策框架之前,先核对工具、渠道、时滞和失灵方式
工具
在判断政策立场之前,先说清具体使用的是哪种工具。同一个目标可以通过利率、支出、税收、监管、沟通或资产负债表操作来实现。
传导
追踪政策变动如何经由家庭、企业、银行、市场、预期、汇率和公共部门资产负债表传导。
时滞
区分宣布、实施、市场反应、实体经济反应以及数据发布的时间。
失灵方式
说明哪些情况会让政策适得其反、生效太晚、效果外溢到国外、只是转移风险,或无法触及受约束的部门。