One month ago, in the aftermath of the FOMC decision which stunned all WSJ-polled economists who were certain a rate hike was imminent by not hiking, instead blaming its on Chinese and global weakness, and when both the market and the credibility of the Fed were about to crack, Goldman did its part to restore the “BTFD” bid when it called, as we previously reported, that no Fed hike would come until at least mid-2016.
A few short days later, as always happens when Goldman makes a contrarian call, this quickly became the new Wall Street mantra, and stocks soared as even more terrible economic news were unveiled.
Then overnight, Goldman’s chief economist Jan Hatzius, who realizes that the only variable that matters for the Fed’s binary decision is where the S&P 500 is trading, and now that the S&P500 is solidly back above 2000 and is fast approaching its all time highs (not to mention is 30 points above Goldman’s year end price target of 2000) says that the possibility of a December rate hike is a substantial 60%, nearly double the Fed Funds futures implied rate of 30-40%, and suggests that yet another volatility risk flaring is in the immediate future, especially if there is even one economic data point in the coming weeks that is not an absolutely disaster.
Of course, if Goldman is wrong, and the economy slips recession and goes straight into depression, then the S&P will not only open limit up, but hit new all time highs before one can say “global thermonuclear war is the best possible news stock bulls can get.“
None of this is news. What is news is that even Goldman dares to take a jab at the Fed’s credibility: to wit:
… the Fed may have developed a credibility problem by failing to follow through on guidance that it never actually provided!
And not only that, but Goldman’s own chief economist dares to call the Fed’s bluff:
A more likely reason for the difference is that some market participants have developed a view that the FOMC is just “constitutionally dovish” and will abandon its current guidance even if the economy does fine in the next two months.
Will Yellen dare to admit that none other than Goldman – which benefits the most from easy money policy – and who is implicitly criticizing the Fed for not only losing credibility but is dictated entirely by asset price levels, is right – not to mention every “tinfoil blog” who has said this for years – and that the Fed is nothing more than a device to preemptively attack any market declines by assuring the BTFDers that the only way to profit in this broken market and depressed economy is to buy each and every market dip on what little faith remains in the US central bank?
Look to Fed’s mouthpiece Jon Hilsenrath for any hints that after pushing the market higher at an epic pace in the past month on nothing bad bad after worse news, that the Fed has had enough of being the topic of derisive laughter among all Wall Street participants.
Below is the full Hatzius note, in the form of a rhetorical Q&A, in which the Goldman chief strategist explains why the time to take profits on the latest “bad news is good news” whiplash has come.
Q&A on Fed Liftoff
- We still expect a rate hike at the December FOMC meeting. The leadership has signaled that such a move is likely if the economy and markets evolve broadly as expected, and our forecast is similar to theirs. However, we are only about 60% confident. Most of the uncertainty relates to the possibility that the economic and market environment—or in a broad sense, “the data”—will be worse than the FOMC’s (and our) expectations.
- The low market-implied probability of a December hike of only 30%-40% probably reflects a mixture of concerns about the data (which we find reasonable) and a belief among some market participants that the FOMC will find an “excuse” to stay on hold even if the economy does fine (which we find unreasonable). The low market-implied probability is not a problem now, but Fed officials will need to find a way to move it much higher by the time of the meeting if they really do want to hike.
- The Fed’s rationale for wanting to start the normalization process is straightforward. In their view, labor market slack has diminished substantially, the link between slack and inflation is stronger than widely believed, and the funds rate is far below the longer-term equilibrium rate so they need to get started well before the economy is back to normal. Consistent with this, even versions of the Taylor 1999 rule that focus on broad labor market slack and assume a low short-run equilibrium funds rate suggest that liftoff is appropriate soon.
- Our own view is that it might make sense to start normalizing in December if we were perfectly confident in our baseline forecast for the economy. But uncertainty around that forecast still argues for waiting longer. The main reason is risk management. At or near the zero bound and with inflation well below target, easing policy in response to a renewed negative shock is both harder and more urgent than tightening in response to a positive shock. This means that there is a greater-than-normal incentive to avoid anything—such as an interest rate increase that ultimately turns out to be unwarranted—that could generate a negative shock.
Today we discuss the possibility of a December funds rate hike in Q&A form.
Q: Why do you still expect the FOMC to hike rates in December?
A: Because the FOMC leadership has said that a rate hike by the end of the year is likely if the economy and markets evolve broadly as expected. Our near-term forecast is similar to theirs, so our baseline is also that they hike.
Q: But didn’t they also signal a hike in the run-up to the September meeting and nevertheless decided to take a pass, despite good economic data?
A: No, the September meeting was very different. First, the committee never clearly signaled a September hike, despite much commentary to the contrary. In fact, our interpretation of Chair Yellen’s June 17 press conference and her July 10 speech was that she had shifted her liftoff expectation from September to December even prior to the turmoil in global markets during August. Second, while the economic numbers between June and September broadly matched expectations, Fed officials have made clear that “data dependence” should be interpreted broadly and also includes shifts in financial conditions that could affect future numbers. Thus, the turmoil sealed the case against a move in September.
Q: Aren’t you overstating the strength of the current signal from the FOMC for the December meeting? After all, Governors Brainard and Tarullo seem to have a very different view from Chair Yellen.
A: True, but this is not a surprise. At the September meeting, 4 out of 17 participants projected no hikes until 2016 or later. Presidents Evans and Kocherlakota had already declared themselves to be part of that group. And once most other participants with past dovish leanings—including Boston Fed President Rosengren—had indicated their support for a 2015 hike, it seemed clear that Brainard and Tarullo were the other two 2016 hikers.
Q: Whether or not it is a surprise, isn’t such open disagreement within the Board of Governors highly unusual historically? And doesn’t it undermine Chair Yellen’s authority?
A: No, we don’t think so. The long-term history of the Board of Governors is not a good guide because the Yellen/Bernanke Fed is a very different institution from the pre-Bernanke Fed. Perhaps because of their backgrounds in academia—a world that prizes open debate—Yellen and Bernanke seem more comfortable with disparate views than their predecessors. And even within the Board of Governors there are instances of open disagreement in recent years that did not undermine the authority of the chair. For example, Governor Warsh wrote an op-ed in the Wall Street Journal just after the “QE2” announcement in November 2010 that was highly critical of the committee’s decision (even though he had not formally dissented at the meeting). Nevertheless, Chairman Bernanke writes in his new book that he was “comfortable” with Warsh’s article at the time.
The introduction of the dot plot in early 2012 has probably further enhanced the spirit of open debate because it forces every participant to write down an explicit funds rate path. Although the dots are technically anonymous, committee members have been moving in the direction of revealing their own policy preferences for some time, well before the recent Brainard and Tarullo comments. This is true not only for regional bank presidents but also for governors. For example, Governor Powell gave an interview shortly after the June 2015 meeting in which he (narrowly) projected a September hike, even though it seemed at that time (at least to us) that Chair Yellen had already moved to a December baseline.
Q: Is the leadership starting to back away from a December hike? New York Fed President Dudley acknowledged on Thursday that the economy was slowing.
A: Dudley did seem a bit less confident on growth—and rightly so, because the recent data really have been worse than expected. Largely because of the weakness in employment, retail sales, and the manufacturing surveys, our current activity indicator (CAI) for September stands at 1.8%, in line with our 1¾% estimate of potential GDP growth. If the economy really is slowing to a trend or sub-trend pace, then liftoff will probably shift into 2016.
But such a conclusion still looks premature. Some of the timeliest indicators such as jobless claims and consumer sentiment in October have been quite strong, and part of the weakness in our CAI may reflect short-term noise, including a big drop in the (very volatile) household employment survey. And while payrolls really were materially softer than expected, the FOMC may have a lower “hurdle rate” for payrolls than we thought previously. In his CNBC interview last week, Dudley said that 120,000-150,000 jobs per month would probably be sufficient to push down the unemployment rate over time. Taken literally, this statement is close to a truism because most economists now probably agree that the “breakeven” pace of job growth is 100,000 or less. But the fact that he used these numbers suggests that the FOMC might view an average payroll growth pace of 150,000 (or perhaps even a bit less) as sufficient to meet its goal of “some further labor market improvement.” This is a slightly smaller number than we would have guessed.
Moreover, it is not just growth that matters. The core CPI rose more than expected in September, and financial conditions have eased significantly since the September meeting. If we take a broad view of whether the environment is surprising on the upside or downside relative to the committee’s expectations, the jury is therefore still out.
Q: What odds would you put on a hike in December?
A: About 60% at this point. Most of the uncertainty relates to the possibility that the economic and market environment will significantly undershoot the FOMC’s expectations. We don’t expect this, but are not very certain. There is also some risk from the renewed fiscal uncertainty in Washington, although our baseline forecast remains that the debt limit will be extended and a government shutdown averted.
Q: So why is the market only pricing 30%-40%? Are others so much more pessimistic about the economic environment than you?
A: No, we don’t think that is the reason. Our near-term economic views are probably fairly similar to the consensus. A more likely reason for the difference is that some market participants have developed a view that the FOMC is just “constitutionally dovish” and will abandon its current guidance even if the economy does fine in the next two months. This view was probably strengthened by the outcome of the September meeting, which went against the predictions of many forecasters; two weeks before the meeting, 80% of economic forecasters were projecting a hike, and even on the eve of the meeting that share still stood near 50%. After the decision, some of these same forecasters then criticized the committee for its misleading communication in the run-up to the meeting. Thus, the Fed may have developed a credibility problem by failing to follow through on guidance that it never actually provided!
Q: Is the low market-implied probability of a hike in itself a reason not to hike?
A: If we are still around 30% on the day of the announcement, then the answer would be yes. We recently showed that the FOMC has a strong revealed preference for seeing rate hike decisions well discounted by the time of the meeting; since 1990, about 90% of all hikes were at least 70% discounted, and at least the last two “first hikes”, in June 1999 and June 2004, were practically 100% discounted. In our view, the FOMC will likely want a decision to hike on December 16 to be largely discounted if and when it occurs.
Of course, there is still plenty of time for the market to change its mind about December, so the low market-implied probability is not a pressing issue at the moment. It is too early for Fed officials to jawbone the market strongly since a lot of the relevant information has yet to be released. But if the economy and markets evolve in a way that is similar to the FOMC’s expectations as of the September meeting but the market is still not pricing a hike after the November employment report released on December 4, then we would expect a strong effort from Fed officials to prepare the market for a hike.
Q: Why does the FOMC even want to hike interest rates? Are they just desperate to get off the zero bound?
A: No. In our view, the “one and done” strategy does not make much sense, and we suspect the FOMC agrees. The reason to hike is a desire to start the interest rate normalization process because the committee thinks it is sufficiently close to its goals. We do not believe that they would hike unless they were at least somewhat confident that they will want to hike again within the next three months.
Q: Do you agree that it makes sense to start normalizing so soon?
A: If we were perfectly confident in our baseline forecast for the economy, it might make sense to start in December.
First, the labor market has improved rapidly, and we recently lowered our estimate of the structural labor force participation rate. As a result, we now think that the remaining amount of slack is no longer all that large. The broad underemployment rate U6 currently stands at 10.0%, about 1 percentage point above our estimate of its full-employment rate; adjusting for the greater volatility of U6, this is equivalent to about a ¾-point gap when translated into headline unemployment rate terms.
Second, our recent research using both regional US data and cross-country data provides some a reasonable amount of support for the Phillips curve. Although the explanatory power of slack alone is still not huge, this does support the Fed’s view that further labor market improvement should eventually push inflation back to the target.
Third, the current funds rate is far below our estimate of the longer-term neutral rate, so it makes sense to lift off from zero some time before the economy is at full employment and inflation back at the target. Consistent with all this, even versions of the Taylor rule that use U6 instead of the headline unemployment rate and build in a depressed short-term equilibrium funds rate suggest that the funds rate should rise above zero soon.
Q: What’s the problem with getting started, then?
A: Uncertainty around our forecasts combined with the fact that erring on the side of hiking too early looks riskier than erring on the side of hiking too late. At or near the zero bound for short-term rates and at a time when inflation is well below target, easing policy in response to a renewed negative shock is both harder and more urgent than tightening in response to a positive shock. This means that there is a greater-than-normal incentive to avoid anything—such as an interest rate increase that ultimately turns out to be unwarranted—that could generate a negative shock. Moreover, financial conditions have already tightened significantly and thereby done much of what the Fed typically hopes to achieve by lifting the funds rate (although some of this tightening has reversed in recent weeks). Finally, while our recent research finds a reasonable amount of support for the Phillips curve, there is still uncertainty about the strength of the link and the most appropriate measure of labor market slack. We therefore think it would be better to wait for confirmation—in the form of at least a modest pickup in wage and price inflation—that the economy is really starting to push up against resource constraints before starting the normalization.
