Close

This page had been redirected to a new URL, please update any bookmarks.

Font Size: A A A

macroblog

« Bad Reporting On Mr. Greenspan's Testimony: Part I | Main | As If The Fed Wasn't Making Enough News Today... »

February 16, 2005

Bad Reporting On Mr. Greenspan's Testimony: Part II

The Financial Times has this little tidbit.

The Federal Reserve on Wednesday took a step closer to setting a formal inflation objective, releasing forecasts that show it expects inflation to remain at 1½-1¾ per cent for the next two years.

The move introduces greater transparency in monetary policymaking by giving a clearer picture of the inflation outlook that guides the central bank's decisions. It may also point to new thinking inside the Fed as the long chairmanship of Alan Greenspan draws to an end.

“They are telling us they have become de facto inflation targeters,” said Angel Ubide, a Fed watcher at a Washington-based hedge fund. “They are telling us that 1½-1¾ is the target, based on their expected policy path.”

These numbers come from the so-called "central tendency" projections contained in the Monetary Policy Report.  Here's the exact wording:

With regard to inflation, FOMC participants project that the chain-type price index for personal consumption expenditures excluding food and energy (core PCE) will increase between 11/2 percent and 13/4 percent both this year and next—about the same as the 1.6 percent increase posted over 2004.

Here's what they said in last July's report:

Core PCE inflation appears to have run a little above an annual rate of 2 percent in the first half of 2004; for 2004 as a whole, most FOMC participants expect it to lie between 1¾ percent and 2 percent. For 2005, the central tendency of the projections for core PCE inflation is 1½ percent to 2 percent.

Here's what they said in the report from July 1996:

Most members of the FOMC expect the rise in the consumer price index over the four quarters of 1996 to be in the range of 3 to 3-1⁄4 percent, about 1⁄4 percentage point higher than they predicted last winter... Assuming no further adverse shocks to food and energy prices, and in the context of the Federal Reserve’s intent to keep trend inflation  well contained, the Committee believes that overall CPI inflation should recede. Accordingly, the central tendency of the FOMC’s forecasts shows CPI
inflation dropping back to the range of 2-3⁄4 to 3 percent in 1997.

In other words, these central tendency projections are contained in every report that accompanies the Chairman's semi-annual testimony.   But for a few details, this report was no different from those that appeared a decade ago (and earlier). Where's the great leap forward in transparency?

There was a slight wrinkle this time.  In the past, the out-year projections were not included until the midyear testimony.  This year the Committee decided to provide a two-year horizon in the year's first report as well.  But nowhere in the presentation is there the slightest hint that this reflects any fundamental change in how the Committee wants the projections to be interpreted. Nor was there any indication of such in Mr. Greenspan's written remarks.

The FT should know better.

February 16, 2005 in Federal Reserve and Monetary Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef00d8343520b553ef

Listed below are links to blogs that reference Bad Reporting On Mr. Greenspan's Testimony: Part II:

Comments

I guess that this shows yet again that the FT (after The Economist) is not the constantly reliable paper that so many expect it to be. But we knew that already. And no, I am not anti-British. My local press - eg French - essentially never reports on macro...

What I think is actually really interesting in the Fed's projection statements - and this could be compared to both the BoE and ECB practice - is the precision of the estimate that they give on a one-year-ahead projection. On the one hand, we have at our disposal the historical standard errors in forecasting (as you know, there is some academic paper on this subject every couple years, plus you surely conduct in-house post-mortems) and on the other hand there is no estimate of confidence interval ever given (as far as I remember, am I wrong ?) in these statements - are they 90% confident or 70% confident etc.. now that would be immensely interesting. I would be really interested in hearing why the Fed doesn't like to give fan charts and so forth.

Posted by: godement | February 17, 2005 at 05:56 AM

godement -

Interesting question. There are those that have suggested we do just that. Of course, with longer range projections we really would begin to enter into the realm of supplying quantitative inflation objectives. We're just no quite there yet.

The decentralized nature of the exercise also makes things somewhat tricky. Each participant currently gets to use whatever methodology they want to contribute these numbers -- I, in fact, have very little insight into how others go about deriving their projections. But, as I think you point out (or at least allude to) in a previous comment, even judgmental forecasts can include a subjective probability distribution.

If we did that now, though, would you know whether differences arise because (a)assumptions aboout the shocks differ; (b) models of the economy differ; or (c) policy assumptions, or preferences over objectives, differ? I wouldn't, although an inflation-report type of presentation could resolve that, I guess. It will be interesting to see how the FOMC settles these types of questions if they eventually decide to move forward -- not a foregone conclusion -- to more ECB- or BOE-like regimes.

Posted by: Dave Altig | February 17, 2005 at 08:06 AM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in