For what it's worth (not much lately), here is Goldman's Jan Hatzius with a Q&A on the Fed's announcement, which now sees the first tapering to start in December, QE to conclude (three months after their prior forecast), and expects the first rate hike to take place in 2016: 8 years after the start of the financial crisis.
From Goldman's Jan Hatzius
- We now expect the QE tapering process to start at the December 2013 FOMC meeting and to conclude in September 2014, three months later than before. Our baseline is that the first step will consist of a $10bn cut in Treasury purchases. These steps remain data dependent in all respects--timing, size, and composition.
- A change in the explicit forward guidance for the federal funds rate is also likely, probably at the same time as the first tapering step. Our baseline is an indication that the 6.5% unemployment threshold is conditional on a forecast of a near-term return of inflation to 2%, and that a lower threshold would apply otherwise. But there are also other options, such as an outright inflation floor or an outright reduction in the unemployment threshold.
- Our forecast for the first hike in the funds rate remains early 2016. The reasons are the large output gap, persistent below-target inflation, and some weight on "optimal control" considerations. The market and most Fed officials see an earlier hike, although the number of FOMC participants projecting the first hike in 2016 rose from 1 to 2 and the "dots"--the projections for the appropriate funds rate at the end of 2016 by FOMC participants--came in below expectations.
Q: When will the FOMC start tapering QE?
A: Although there is uncertainty in both directions, we now expect the QE tapering process to start at the December 2013 FOMC meeting. On the one hand, the committee hinted that it is still considering a tapering move soon, by saying that "… [i]n judging when to moderate the pace of asset purchases, the Committee will, at its coming meetings, assess whether incoming information continues to support the Committee's expectation of ongoing improvement in labor market conditions and inflation moving back toward its longer-run objective" (our emphasis). On the other hand, there are two good reasons to believe that a December move is a more natural expectation than an October move.
First, the chairman emphasized the committee's desire to assess how the economy is holding up following the increase in mortgage rates and the tightening of financial conditions. This is important because the amount of economic data to be released by the October 29-30 meeting is limited to just one set of most of the major monthly economic reports, including employment, ISM, retail sales, CPI, and housing starts. The third-quarter GDP report will be released on the morning of the second day of the meeting, but this may be too close to decision time to have a decisive effect and, in any case, looks relatively soft according to our tracking estimate of 1.8%.
Second, while a scheduled press conference and an updated Summary of Economic Projections (SEP) is perhaps not essential for a move, as the Chairman indicated, it is certainly helpful in order to put the decision to taper QE into perspective. In our view, a hastily arranged conference call is only an imperfect substitute. This is a particularly important point as the Chairman confirmed today that the committee is considering changes to the forward guidance for the federal funds rate. To the extent that the committee views the SEP as an important aspect of the forward guidance, it may want to wait for an SEP meeting before tapering and changing the guidance.
Q: What will the first step look like?
A: Our working assumption is a $10bn step-down in the pace of Treasury purchases, with no change in MBS purchases. There was little new information about the size of the first step or its composition, but the committee's evident trepidation about the impact of tapering argues for a small step that is concentrated in the asset class widely believed to have less impact on economic activity, namely Treasuries.
Q: When will the QE program end?
A: We have pushed back our expectation for the end of the program by three months to September 2014. This change is based on two considerations.
First, stating the obvious, a later start date for the tapering process implies a later end date unless the committee increases the pace of tapering. If there is a $10bn cut in December, the committee would need to step up the tapering pace to nearly $20bn per meeting to get to zero by midyear. This is possible, but seems too aggressive as a central expectation.
Second, Chairman Bernanke backed away from his prior guidance that the program would end in June 2014 with an unemployment rate of 7%. Instead, he emphasized that "there is no fixed calendar" and that "the unemployment rate is not necessarily a great measure in all circumstances." We interpret his remarks as a hint that the program is now more likely to end in the second half of 2014 and after the unemployment rate hits 7%.
Q: Will the committee change the forward guidance for the federal funds rate?
A: This seems likely, and we still believe that it would probably coincide with the first tapering step. In fact, while the committee declined to change the "mix of instruments" at today's meeting, Chairman Bernanke provided his clearest argument to date for such a move in the future: "We have asset purchases, and we have rate policy and guidance about rates. It’s our view that the latter, the rate policy, is actually the stronger, more reliable tool."
However, the form of this change is unclear. Our baseline expectation is that the committee will indicate that the 6.5% unemployment threshold is conditional on a forecast of a near-term return of inflation to 2%. This would mean that if inflation was still meaningfully below 2% by the time the 6.5% threshold was reached, the committee would not regard the threshold as having been met until the unemployment rate had fallen to a lower number (which may or may not be explicitly specified).
However, other changes in the guidance are also conceivable. First, in response to a question from Greg Ip of The Economist, Bernanke referenced the concept of an "inflation floor" as one possibility. For example, the committee could say that it would not raise interest rates if inflation were expected to remain below a certain threshold over a particular period. However, such a floor is not without its problems. On the one hand, it might not provide much additional information if the threshold is very low or the time period for the forecast period is very long. (At a sufficiently long time horizon, Fed officials will always project that inflation will converge to their target.) On the other hand, it might expose the committee to the risk that the economy could overheat materially, or a large asset bubble could develop, if there was some factor other than resource utilization--e.g., a sharp and sustained expansion in commodity supply that pushed down commodity prices over a lengthy period--that held inflation at very low levels for a time. One example is the late 1990s, when the unemployment rate fell to 3.8% but inflation remained persistently below 2%.
Second, the committee might lower the 6.5% unemployment threshold on an outright basis, as discussed in the minutes of the July meeting. However, in response to a question, the Chairman declined to say whether the committee had discussed this option today, merely repeating his reference to an inflation floor and noting that "there are a number of options that we have talked about."
Q: So when will the committee actually hike the funds rate?
A: Our forecast remains that the first hike will occur in early 2016. This is significantly later than the market consensus. It is also later than the projections of appropriate policy by 15 of the 17 FOMC meeting participants, although the number of participants in the 2016 camp rose from 1 in June to 2 in September. There are three reasons for our dovish view.
First, we believe there is significantly more slack in the labor market than indicated by the unemployment rate alone. At present, we estimate that overall employment is 3.5%-4% below its potential level, taking into account both the (estimated) gaps between the actual and structural unemployment rate and the actual and structural labor force participation rate.
Second, and related to this, we expect inflation to remain below the committee's 2% target through 2016. We would note that the committee leans in the same direction; this is particularly true for the headline PCE index, where the SEP central tendency is 1.7%-2.0% for 2016.
Third, "optimal control" considerations argue for delaying the timing of the first rate hike relative to the implications of simple policy rules. This could be particularly important if Vice Chair Yellen ascends to the Chairmanship early next year, as she has spoken extensively about the importance of this approach in the past.