From Goldman's Andrew Tilton:
Summary of tomorrow's events:
- Wednesday afternoon will mark a watershed in Federal Reserve communications strategy. Chairman Bernanke will give his first post-meeting press conference and the Federal Open Market Committee’s “central tendency” economic forecasts will be released at the same time (rather than three weeks later with the minutes, as has been the practice recently).
- We expect 1) modest revisions to the FOMC’s forecasts to reflect recent news of softer Q1 growth and higher inflation, 2) reiteration of the intent to end QE2 in June, but an indication that reinvestment of maturing securities is likely to continue beyond that time, 3) a relatively dovish tone from Chairman Bernanke in the press conference, given the still-high level of unemployment and the many uncertainties in the growth outlook
Wednesday afternoon will mark a watershed in Federal Reserve communications strategy. In addition to the Federal Open Market Committee (FOMC) policy statement, Chairman Bernanke will give his first post-meeting press conference and the FOMC's “central tendency” economic forecasts will be released. The expected timeline of events is as follows:
12:30pm – FOMC statement released. The Fed’s website gives the time as “around 12:30pm”, which will come as no surprise to Fed watchers used to twiddling their thumbs for several minutes after the scheduled release time. (Just to keep market participants on their toes, the statement does occasionally come out a minute or two before the scheduled time.)
2:15pm – Press conference begins. We expect Fed Chairman Bernanke to make an introductory statement which will feature the FOMC’s projections for growth, unemployment and inflation (but probably not the detailed distribution of these forecasts nor discussion of the staff’s forecasts). An article published today on the Wall Street Journal website (“Federal Reserve Irons Out Details of Post-Meeting Press Conference”, by Jon Hilsenrath) implied that any introduction is likely to be very short. According to the article, additional published information on the FOMC forecasts, along the lines of Table 1 in the Fed minutes from the January 25-26 meeting will be made available on the Fed website at this time. Assuming this is correct, it would imply that detailed information on the distribution of forecasts, and on the staff’s economic forecasts, would not be revealed until the publication of the minutes (though of course these subjects could surface during the question and answer session).
ca. 2:25pm – Question and answer session begins. This is a live session with journalists, who are likely to be well prepared with probing questions. The questions have not been submitted to the Fed in advance.
Sometime around 3pm or slightly after – Press conference ends. The typical length of an ECB press conference in recent years has been about 45 minutes, perhaps a little longer recently (see below). The aforementioned WSJ article suggested a similar length for the Fed’s first conference.
Wednesday’s events are of importance to market participants for at least three reasons, in our view: 1) the information they will provide about the FOMC’s economic outlook, 2) communication of decisions regarding the asset purchase program (“QE2”), and 3) the novelty of the press conference format itself, which should provide at least some insight into the Fed Chairman’s thinking (although he will undoubtedly emphasize the views of the committee rather than his own).
The FOMC’s Economic Outlook
Both the description of the economic outlook in the Fed statement and the formal projections of FOMC members will likely reflect recent news of softer Q1 growth and higher inflation. Although it’s not clear that the recovery is “on a firmer footing” than it was in March, as the lead sentence of the statement implies, we expect the economic writeup in the first two paragraphs of the March 15 policy statement to survive largely intact. The description of the various components of GDP – consumer spending and business equipment and investment “continues to expand”, structures investment “is still weak”, and the housing sector “depressed” – remain accurate. The labor market is clearly improving, so the “gradually” qualifier might be dropped. Commodity inflation remains a problem, but long-term inflation expectations remain stable as well. The unemployment rate is still “elevated” and underlying inflation “somewhat low” relative to mandate-consistent levels.
In terms of numbers, the Blue Chip survey of economic forecasters shows a small decline in 2011 real GDP forecasts (Q4/Q4 basis) from 3.3% prior to the FOMC’s January forecasts to 3.2% in early April. Meanwhile, forecasts for (headline) consumer price inflation in 2011 rose from 1.8% to 2.9%. We expect revisions to the FOMC’s economic projections to follow a similar pattern. The table below displays the last set of forecasts (compiled for the January 25-26 meeting) and our expectations for the numbers to be released Wednesday.
The FOMC's January Economic Projections, and Our Expectations for the April Projections
Source: Federal Reserve. GS Global ECS Research.
The Policy Decision
The immediate policy decision at hand is what to signal with respect to the Fed’s plans for further securities purchases. At the last meeting, the FOMC stated it was “maintaining its existing policy of reinvesting principal payments from its securities holdings and intends to purchase $600 billion of longer-term Treasury securities by the end of the second quarter of 2011”. This wording is an invitation—which we accept—to think about the asset purchase program and principal reinvestment separately.
As for the asset purchase program, all signs point towards conclusion on schedule in June. We see no chance of an expansion of the program at this time, as the political backlash following the announcement of “QE2” was significant, and no one on the FOMC has publicly made the case for an expansion. Likewise we see essentially no chance that the FOMC will end the program early, as this could be seen by some as an implicit admission that it was not necessary—an implication which we think most Fed officials do not believe or would want to convey.) The path of least resistance is clearly to end it on schedule, and recent speeches by key FOMC members are consistent with that course of action. Media reports over the last few days also suggest that it will end on schedule. We do not expect the Fed to taper its purchases in the final weeks of the program, and recent press reports have reinforced our view; our own research implies that whether or not the Fed does so would have little impact on markets (see “QE Progress Report,” Zach Pandl, US Daily, March 24, and “Stocks vs. Flows Revisited: End of QE2 Unlikely to Have Significant Effect on Bond Yields, Jari Stehn, US Daily, April 13, for more on the tapering issue and the impact of QE2 on markets).
The reinvestment of Fed purchases is less important for its asset market effects – reinvestment amounts in the second half of the year will probably be on the order of $20bn per month, a small fraction of the purchase rate currently – than for the signal it sends about future monetary policy. Continuing the reinvestment policy would imply a flat balance sheet in coming months – i.e. neither easing nor tightening. (Board of Governors Vice Chair Janet Yellen gave a speech in February that cited a staff “illustrative baseline scenario” where the balance sheet remained flat for some time after mid-2011. See “Unconventional Monetary Policy and Fed Communications,” February 25.) Ending reinvestment would mean tightening on the margin, albeit extremely gradually, and would immediately raise the question of when the next tightening step (presumably a change in the forward-looking language in the policy statement) might occur. Given the uncertainties about the growth outlook, this is probably not a message Fed officials want to send at the moment. We expect reinvestment to continue, but without any guidance about how long it might go on.
The Press Conference
With Wednesday’s press conference, the Fed joins a number of other central banks – including the European Central Bank and the central banks of Switzerland, Norway, Sweden, the Czech Republic, Poland, New Zealand, and Japan – that have adopted this method of communication with the markets and the public. Research on central bank communication suggests that a) the press conference can convey significant market-moving information beyond the policy decision itself, b) markets tend to move in the same direction during the press conference that they did in reaction to the statement, c) the most important statements during the press conference are those relating to inflationary developments and the committee’s discussion of rate policies. (See “Explaining Monetary Policy in Press Conferences”, Michael Ehrmann and Marcel Fratzscher, International Journal of Central Banking, June 2009, pp. 41-84. Note that the ECB, which was the focus of this analysis, releases only a rate decision and not a policy statement before the press conference, which leaves more information to be revealed at the press conference.)
The press conference presents both risks and opportunities for Chairman Bernanke and the Fed. Although similar in some respects to Congressional testimony, press conferences are likely to feature shorter and perhaps in some cases more technical questions. The academic paper cited in the previous paragraph found that the average press conference over the 2001-2006 period lasted 44 minutes and featured 16 questions; in recent years the conferences seem to have been a little longer (indeed, the last press conference featured 23 questions, of which quite a few were two-part questions). Media coverage, at least for the first press conference, is likely to be intense. Bernanke will be speaking to the public and trying to instill confidence in the Fed’s policies—both conventional and “unconventional”—so demeanor or “body language” may matter as much for the perceived success of the press conferences as the technical precision of his answers.
We see two sets of implications, one fairly straightforward and the other more speculative. It seems clear that press conferences will 1) increase Fed Chairman Bernanke’s ability to control the message relative to the rest of the committee, as he becomes the first and premier interpreter of the Fed statement, 2) decrease the importance of the policy statement itself (on days when a press conference follows), as there is less reason to guess the meaning of vague phrases such as “somewhat weaker” when hard numbers on the FOMC forecasts are to follow, 3) potentially increase the market impact of the FOMC’s forecasts, as they will be timelier when first released, 4) decrease the importance of the minutes (again, following meetings with a press conference), since markets will already have both key details of the FOMC forecasts and color on the meeting discussion, 5) introduce more market volume and volatility during press conferences.
More speculatively, press conferences could 1) reduce the likelihood of a dissent being initiated at the associated meeting, as these will be more likely to put the Chairman in an awkward position and will also be subject to his interpretation without opportunity for an immediate rejoinder by the dissenter, 2) reduce the market’s focus on speeches by other FOMC participants in the immediate aftermath of press conferences, with market interest increasing as the next scheduled press conference approaches, 3) alter the content or timing of speeches or other commentary by other FOMC participants, who might desire or be asked to comment on the Chairman’s interpretations, or alternatively take pains to avoid contradicting the Chairman’s statements from the press conference, 4) create considerable headline risk for future Fed Chairs, as they will conduct their first few press conferences without the extensive experience Bernanke now has with Congressional testimony and media interviews.
During the press conference, Bernanke will likely focus on characterizing the committee’s views rather than his own. The discussion of the FOMC’s central tendency forecasts is an obvious place to begin, and the apparent decision not to release the distribution of individual forecasts until the minutes will contribute to the focus on areas of agreement rather than disagreement.