央行应该提供货币政策时间表吗？

金融危机过后，利率仍保持为零，央行试图用各种新举措来刺激经济。政策之一是前瞻指引：承诺将货币宽松政策维持一段时间。美联储在2011年表示，低利率将持续至2013年中，随后又延续到2014年，接着推迟到2015年。前瞻指引现已成为货币政策的工具之一。一旦利率上升，如果使用这个工具才能达到最佳效果呢？上周，纽约芝加哥布斯的货币政策论坛发布了一篇文章，认为在正常情况下，央行的承诺不一定有用。 

前瞻指引有两种，其一为时间表。美联储刚刚在十月采用了这一方式，他们在会后声明中强调了下一次会议的重要性，而美联储确实在十二月的会议中宣布加息。另一种方式是不公布具体时间，只说明总体方针，指明形势向不同方向转变时利率将如何变化。举例来讲，英格兰银行在2014年承诺，在失业率降至7%之前，不会加息。Michael Feroli、David Greenlaw、Peter Hooper、Frederic Mishkin 以及 Amir Sufi 认为，由数据驱动的指引要强于时间表。数据驱动的指引能够清晰地表明央行的态度和倾向，它的好处有两点。首先，如果市场事先了解美联储修改利率的计划，那么利率变化之时，投资者们就不会乱发脾气了。在美联储十二月加息之时，加息的消息早已众所周知，这意味着市场不会出现大幅波动。其次，如果利率紧随金融态势发生变化，货币政策就具备了自动转向的能力，在美联储采取措施之前，货币政策已经对事件做好了反应。 

既然提到了政策时间表，就不得不提及成本问题。首先，时间表会同数据驱动的指引发生冲突。财经媒体和交易员在关注汇率变化时间表的同时，会忽略与之相伴的预警。更严重的情况，时间表式的指引会与数据驱动的指引大相径庭，会让市场误以为美联储的政策只会按计划进行，而忽略了实际的经济形势。这样的货币政策是事先制定好的，不能像数据驱动的指引一样进行自我修正。事实证明，美联储越强调加息时间，投资者们越容易忽略财经消息。

但在利率维持在较低水平时，时间表还是比较有时效性的。刺激经济增长并摆脱流动性陷阱的一种方法就是承诺未来的通胀率会增加。央行做出如此承诺却很难让人信服，因为他们一旦逃离了流动性陷阱，立刻会对激增的通胀进行打压。除非他们敢承诺在经济复苏之后仍保持宽松的货币政策。在会议中，旧金山联储主席和前任利率设定者一致认为，在危机后的通信中，美联储该做的是指出一条明确的出路，而不是讨论哪条路更好。 

正常时期的通信也很让人担心。评论员过分关注于时间表，就连英格兰银行7%的失业率阀值都被市场和媒体忽视了。这意味着时间表是无可避免的。除此之外，时间表简单易懂，不同形势下采取的不同措施却很难解释，尤其是在官员们都各持意见难以说清的时候。就算看在时间表简单易懂的份上，央行也不能将之随意抛弃。 

Should central banks offer timelines for monetary policy? 

After the financial crisis, with interest rates stuck at zero, central banks tried to gee up economies with a variety of new measures. One was forward guidance: pledging to keep policy loose for a while. In 2011 the Federal Reserve said it expected rates to remain low until mid-2013; this was then pushed back to mid-2014, then mid-2015. Forward guidance is now an established part of the monetary policy toolkit. But what is the best way to deploy it once interest rates are rising? A new paper, presented last week at Chicago Booth’s Monetary Policy Forum in New York, argues that in normal times, not all central bank promises are equal. 

There are two types of forward guidance. One uses a timeframe. The Fed did this most recently in October when, in its post-meeting statement, it referred explicitly to the importance of the next meeting (in December, when it raised rates). The alternative is not to mention dates, but say how, in general, interest rates would respond to different turns of events. For example, in 2014 the Bank of England promised—with some caveats—not to raise rates until unemployment fell below 7%.

Michael Feroli, David Greenlaw, Peter Hooper, Frederic Mishkin and Amir Sufi argue that in normal times, data-dependent guidance beats a schedule. Data-based guidance helps mainly by demystifying the central bank’s preferences. That is beneficial for two reasons. First, if markets know how the Fed will change rates when the economy soars or sours, they will not throw a tantrum when rates change. When the Fed raised rates in December, its move was widely expected. That meant markets barely moved. Second, if market interest rates respond immediately to economic news, monetary policy benefits from automatic steering; financial conditions automatically loosen and tighten in response to events, before the Fed actually turns the steering wheel. 

But mentioning timelines for policy, the authors argue, comes with costs. First, timelines distract from data-driven guidance. The financial press and traders focus on any hint of a schedule for changes in interest rates, and often ignoring the accompanying caveat. Worse, time-based guidance can work against data-based guidance, by leading markets to think that the Fed will stubbornly stick to its plan whatever happens to the economy. Monetary policy is set on cruise control but, unlike with data-driven guidance, it does not steer itself. The authors present evidence that market interest rates become less sensitive to economic news the more the Fed mentions timelines or dates in its statements.

Timelines may remain sensible, though, when interest rates are at their lower bound. One way to stimulate the economy and escape a liquidity trap is to promise higher inflation in the future. It is tough for central banks to make such promises credibly; having escaped the liquidity trap, their instinct will be to clamp down on resulting inflation. It might therefore be helpful if they can tie their own hands, promising to keep policy loose even after the recovery. At the conference both John Williams, president of the San Francisco Fed, and Jeremy Stein, a former rate-setter, suggested that post-crisis Fed communications were partly an attempt to commit to a given path, rather than merely to communicate preferences. 

The authors, though, are concerned with communication in more normal times. And they have a point about commentators’ obsession with timelines; even the Bank of England’s 7% unemployment threshold was immediately interpreted by markets and the press as providing a timeline for a rate rise, when paired with the Bank’s unemployment forecast. That suggests, though, that it is tough to avoid timelines altogether. And schedules come with the virtue of simplicity; guarded statements about possible different paths for rates can be tricky to interpret, especially when they are the work of a committee whose members’ individual views differ. Central banks should not throw out the planner entirely—if only for simplicity’s sake. 

本文翻译由兄弟财经提供

文章来源：http://www.economist.com/blogs/freeexchange/2016/03/dates-vs-data