Fed Wrestles With Setting Inflation Target Opinion Moves Toward a Flexible 'Objective'
That Gives Weight to Unemployment
WASHINGTON -- One of Federal Reserve Chairman Ben Bernanke's top priorities is to improve the central bank's communication with the public, especially about its goals for inflation and growth and how it plans to achieve them. But the more he and his colleagues concentrate on that task, the harder it appears.
Fed officials spent much of last week's meeting discussing communications, a topic that has been the focus of a subcommittee headed by Vice Chairman Donald Kohn, a 36-year Fed veteran. Minutes of that meeting, due out Nov. 15, may offer clues to the discussion. There has been sentiment inside the Fed for change, and some sentiment for unveiling any changes as a package rather than one by one. But the complexity of the issue suggests it could be months before officials agree to anything.
Under former Chairman Alan Greenspan, the Fed resisted the international trend of declaring a numerical inflation target as a way of fostering clarity and accountability for its goals. With Mr. Bernanke, a longtime advocate of numerical targets, at the helm, opinion within the Fed's policy-making Federal Open Market Committee is moving in favor of some kind of numerical "objective." The FOMC is composed of the Fed's six current governors -- there is one vacancy -- and presidents of its 12 regional banks.
Target advocates, including Mr. Bernanke, have moved toward a flexible target that gives greater weight to restraining unemployment, not just prices. In fact, they prefer to use the softer term, "objective." Meanwhile, Fed officials who have been skeptical of targets increasingly see that their deliberations can be hurt when members are advocating different interest-rate policies because they have different inflation goals.
Agreeing in principle on a target is just the first step, however. Officials would have to pick a number, a yardstick and time period in which to hit their goal. In a 2002 congressional hearing, Mr. Bernanke suggested an inflation target of 1% to 2% using the "core" price index of personal consumption expenditures, or PCE, to be achieved "over the medium term, say one to two years." The "core" index excludes the volatile food and energy sectors, which can mask the underlying inflation trend.
Since then, eight of the FOMC's current 18 members have advocated that range or something close to it. But 1% to 2% may not be the preferred target. Gary Stern, president of the Federal Reserve Bank of Minneapolis, says, "There seems to be something of a public perception that that's a semiofficial range, and from my perspective there's nothing remotely official about it."
Indeed, the closer some officials look, the more they believe a 1%-to-2% range is too low. At 1% inflation, there is a greater risk a shock could tip the economy into deflation, or generally falling prices. Because nominal interest rates can't go below zero, an inflation rate of 1% or less gives the Fed little room to make the kind of cuts in inflation-adjusted interest rates that might be needed to remedy severe economic weakness.
Research by Fed economist Athanasios Orphanides, Günter Coenen of the European Central Bank, and Volker Wieland of Goethe University Frankfurt concludes the Fed would encounter this "zero nominal bound" on interest rates once every 20 years with a 2% target; every 10 years with a 1% target; and every five years with a target of zero. This implies "a significant deterioration" in economic performance with an inflation target of 1% or lower, they conclude. Their research finds these risks are insignificant with a 2% target. The ECB's stated target is "below, but close to, 2%."
Some Fed officials are reluctant to voice misgivings about the 1% to 2% range for fear of giving the impression they are moving the goal posts while the Fed is trying to nudge inflation lower. That concern may be unnecessary: Surveys, and yields on inflation-indexed bonds, suggest investors and economists perceive the Fed's inflation goal to be 2% to 2.5%.
The price index that would be used is also uncertain. For years the Fed has preferred the core PCE price index, arguing that its composition better reflects consumer spending patterns than does the popular consumer-price index. Measured by the PCE, inflation generally runs about half a percentage point lower than with the CPI.
But the case for using the PCE as a target is no longer clear. In recent years, the PCE has been subject to large revisions. The CPI generally isn't revised except for seasonal adjustment. Many of the features that led it to overstate inflation have been corrected. And it is the inflation measure most familiar to the public.
"Some of my earlier concerns about the core CPI have diminished, in part because my concerns about the core PCE have increased," says Mr. Stern of the Minneapolis Fed.
Some foreign central bankers fault the Fed for focusing on an inflation indicator that excludes food and energy prices. Because energy prices have risen steadily in recent years, they say core inflation has consistently understated actual inflation. But so far, Fed officials appear content with the core indicator. "Historically, headline and core inflation rates have the same average....The past three years have been anomalous," says St. Louis Fed President William Poole.
Advocates of inflation targets usually argue that a central bank should give a timetable to achieve its target to avoid the temptation to delay action when inflation is above the target. "If a target is going to be meaningful, it will have to have a horizon associated with it," Mr. Stern says. Yet many of his colleagues argue for flexibility to cope with a shock, such as an oil-price surge. A rigid schedule might require sharply higher interest rates, driving up unemployment and compromising the Fed's duty to restrain inflation and unemployment.
"I favor an elastic 'medium term' specification for an inflation target, rather than a specific period such as two years," Mr. Poole says.
Another thorny issue is how detailed an inflation forecast to make public, something generally seen as critical to explaining how the Fed expects to achieve its goals. The Fed currently releases a summary of all the FOMC officials' forecasts; but because those officials may be using differing assumptions for interest rates, oil prices and other factors, the summary of their views may not represent a coherent view, limiting the summary's usefulness. By contrast, the nine members of the Bank of England's policy committee, all monetary-policy experts, release a single forecast with specific assumptions.
The British approach, however, might not work in the U.S., some officials say, because the FOMC is larger, more dispersed, and some of its members aren't experienced forecasters.
Fed在设定通胀目标上犹豫不决
美国联邦储备委员会(Fed)主席本?贝南克(Ben Bernanke)的首要任务之一是加强Fed与公众的沟通，特别是在有关该机构的通货膨胀和经济增长目标以及它计划如何实现这些目标方面。但贝南克及其同僚越上心，这项工作似乎却越难做。
Fed官员们在上周会议的大部分时间都在讨论沟通问题，这一直是Fed副主席唐纳德?科恩(Donald Kohn)所领导的一个小组委员会重点关注的问题。定于11月15日公布的这次会议的纪要可能会为人们了解Fed官员的相关讨论提供线索。Fed内部一直存在要求改革其对外沟通工作的呼声，还有Fed人士希望这方面的改革能够毕其功于一役，不要零敲碎打地进行。但此事的复杂性表明，近几个月Fed官员可能还无法在这一问题上形成任何共识。
在格林斯潘(Alan Greenspan)担任Fed主席期间，该机构一直拒绝顺应设立具体通胀率目标的国际潮流，这种国际流行的做法是为了提高央行政策目标的明确性和可信性。鉴于一贯支持对政策目标进行量化的贝南克现在成了Fed主席，在Fed决策机构──联邦公开市场委员会(FOMC)内部，支持设立某种“量化目标”的意见正逐渐占据上风。FOMC目前由6位Fed现任理事(另有一个名额空缺)及美国12个地区性联邦储备银行行长组成。
包括贝南克在内的量化目标支持者倾向于设立一个柔性的通货膨胀率目标，该目标更偏重控制失业率，而不仅仅是维持物价稳定。与此同时，那些一直对设立量化目标持怀疑态度的Fed官员们日渐认识到，由于Fed成员心目中的通货膨胀率目标各不相同，这就造成了他们在利率政策上各持己见的客观现实，从而对Fed的决策造成损害。
不过，在设立通货膨胀率目标问题上原则上达成一致仅仅是迈出了第一步。Fed官员还需确立具体的目标数字并设定在多长时间内实现这一目标。贝南克在2002年举行的一次国会听证会上建议，以“核心”个人消费支出价格指数(PCE)增幅“中期内，即1到2年内”降至1%-2%的水平作为通货膨胀目标。这一“核心”价格指数不包括波动性较大的食品和能源的价格，这两类价格会导致核心通货膨胀趋势失真。
自那以来，FOMC目前18个成员中的8人都表示支持这一目标或与其接近的目标值。但1%-2%的区间可能并非Fed的首选目标。明尼阿波利斯联邦储备银行行长加里?斯特恩(Gary Stern)说：“公众或许认为这是一个半官方的通货膨胀率目标区间，但我个人认为，这个指标根本没有多少官方性质。”
事实上，一些Fed官员越研究越发现，1%-2%这一通货膨胀率区间太低了。如果通货膨胀率只有1%，那么很有可能一有风吹草动美国经济就会陷入通货紧缩，即物价总体呈下降之势。由于名义利率不可能低于零，如果通货膨胀率等于或低于1%，那么Fed在下调经通货膨胀因素调整后的利率方面就没有多少可施展手脚的余地了，而降息是刺激经济增长的一个重要手段。
Fed的经济学家阿萨纳斯奥斯?欧菲尼德斯(Athanasios Orphanides)、欧洲央行的甘特?可尼(Gunter Coenen)和法兰克福歌德大学的沃尔克?魏兰(Volker Wieland)通过研究认为，如果Fed将目标通货膨胀率定在2%的水平，那么每隔20年就会出现一次“零名义利率现象”；如果定在1%，那么每隔10年就会出现一次“零名义利率现象”；如果将目标通货膨胀率定为零，那么每5年就会出现一次这种现象。这意味着，如果要将目标通货膨胀率定在1%以下，那么经济表现将会“严重恶化”。他们的研究发现，如果将通货膨胀率目标定为2%，那么这种风险将不足为虑。欧洲央行因此将其目标通货膨胀率定在“低于但接近2%的水平”。
一些Fed官员不愿质疑将目标通货膨胀率设在1%-2%的合理性，怕给外界留下他们在与正努力降低通货膨胀率的Fed唱反调的印象。这种担心或许是多余的：市场调查和通胀指数债券的收益率都显示，投资者和经济学家们都认为Fed的目标通货膨胀率是2%至2.5%。
用哪一个物价指数来衡量通货膨胀率也还是个未定之事。Fed多年来一直倾向于核心的PCE价格指数，认为它比常用的消费价格指数(CPI)更能反映人们的消费支出曲线。以PCE反映的通货膨胀率通常要比CPI反映的通货膨胀率低0.5个百分点左右。
但以PCE指标来确定目标通货膨胀率也不再那么十拿九稳了。近年来，人们往往要对PCE指标作大幅修正。而CPI指标除季节性因素外一般不会因其他因素而作修正。许多导致CPI指标夸大通货膨胀现象的方面都已得到了纠正。而CPI也是公众最熟悉的通货膨胀指标。
明尼阿波里斯联邦储备银行行长斯特恩说：“我较早时对核心CPI指标的担忧现在已有所减弱，这部分是因为我对核心PCE指标的担忧增加了。”
其他国家的一些央行官员认为Fed只关注不包括食品和能源的核心通货膨胀指标是不对的。他们说，鉴于能源价格近年来一直在稳步上涨，因此核心通货膨胀指标一直低估了真实的通货膨胀状况。但目前为止Fed官员们似乎仍对核心通货膨胀指标信任有加。圣路易斯联邦储备银行行长威廉?波尔(William Poole)说：“从历史上看，总体通货膨胀率与核心通货膨胀率所反映的趋势是一致的，过去3年的情况是个例外。”
支持设立目标通货膨胀率的人士声称，央行应该定出一个达到这一目标的时间表，以免在通货膨胀率高于这一目标水平时延误行动时机。斯特恩说，如果要使目标通货膨胀率有意义，就应该为它规定一个达到的时限。但他的许多同僚却认为，这方面应灵活掌握，以便为油价暴涨等突发性因素留有应对的余地。僵硬的时间表有可能要求Fed在不适当的时候大幅上调利率，从而推高失业率，使它无法同时兼顾抑制通货膨胀和确保就业的双重使命。
波尔说，他倾向于“中期内达到目标通货膨胀率”这一比较有弹性的说法，而不喜欢将时间定得太具体，比如说两年。
另一个棘手问题是，Fed对通货膨胀前景所作预测应对外公开到何种程度，人们普遍认为，对外公开的程度能表明Fed实现通货膨胀目标的决心大小。Fed目前会发布一份逐一介绍每位FOMC成员所作预测的报告，但由于这些官员在作预测时所假设的利率、油价和其他指标各不相同，因此人们可能无法从这份报告中总结出Fed的总体看法，从而限制了这份报告的有效使用。相比之下，英国央行货币政策委员会公布的通货膨胀前景预测报告就是总结了其9位成员观点的单一预测结果，预测中所假设的各种指标也是统一的。
但一些官员认为，英国的这种做法可能不适用于美国，因为FOMC的成员更多，他们的意见更难统一，而且其中一些人并不擅长预测。
Greg Ip
