Tim Duy's Fed Watch

The Final Countdown

The latest read on the Fed’s preferred inflation metric was not particularly kind to policymakers:

  

Indeed, as Craig Torres at Bloomberg notes, this is only one of a number of indicators that should give a reality check to FOMC participants as December’s meeting approaches. A stronger dollar, weaker commodity prices, and falling inflation expectations suggest that the “transitory” negative weights on inflation might persist longer than the Fed anticipates.  

In addition, since I last wrote, real time estimates of fourth quarter GDP weakened in the face of incoming data. And manufacturing is data is off to a weak start this month with a fall in the Chicago PMI. Indeed, manufacturing indicators are weaker than we would normally see at the onset of a tightening cycle, but the Fed is betting that these indicators are passé in a world dominated by services. And that side of the economy seems to be holding up nicely:

We get fresh national readings from the Institute of Supply Management this week. Still, even if the numbers are on the soft side, there is little I think that will dissuade officials from hiking rates in December. With unemployment at 5% and wage growth picking up to confirm receding slack in the labor market, the general consensus on Constitution Avenue is that the time is ripe to nudge rates higher. Wait any longer, the thinking goes, you risk being unable to raise rates “gradually.” It will be interesting to see how the Fed would react to a weak November labor report, due Friday. It seems as long as the employment report is not a complete disaster, even numbers on the soft side would be enough to justify Fed action on the basis that the underlying trends remain in place.

As Torres also notes, even if December is pretty much in the bag despite questions about inflation, the path of subsequent rate hikes will depend on confidence in the path of inflation. Boston Federal Reserve President Eric Rosengren via an extensive interview with the FT:

My own personal view is we should have a flexible approach to thinking about the path with gradual being the important consideration, but we are still not near 2 per cent inflation. By the core PCE at 1.3 per cent we are still pretty far away. What gives me reasonable confidence about the path of inflation is the fact that the labour market slack seems to be diminishing relatively quickly. But I would want to continue to see progress on wages and prices moving up. If we weren’t seeing wages and prices moving up over time our willingness to keep raising rates would go down . . . 

So it is partly conditional on whether that reasonable confidence, as your rates get higher you should probably want a standard that is a little higher than reasonable confidence. I would not expect to continue to see 1.3s for the core PCE. If we continue to see 1.3 [per cent] for the core PCE we would have to think about why is inflation not picking up towards our 2 per cent goal.

Will wages and prices move higher? I would be surprised if this wasn’t the case, assuming of course the economy maintains sufficient cyclical momentum to sustain further improvements in the labor force. An often-overlooked point is that wage growth is arguably not as puzzling as it seems. Consider the Atlanta Fed Wage Growth Tracker, which estimates median wage growth of matched individuals, those with earnings now and twelve months ago. By tracking persons with continuous employment over a year, this metric avoids the problem of compositional effects due, for example, to persons entering and leaving the labor force due, for example, to demographic shifts or cyclical factors. The Atlanta Fed measure compared to other measures of wages:

The Atlanta Fed measure accelerated in 2015 as unemployment moved below 6%:

Note the deceleration in wages in recent months; this is attributable to lower wage growth for women. Looking at men only, the relationship between unemployment and wage growth is somewhat tighter:

  

The Atlanta Wage Growth Tracker suggests that the underlying relationship between unemployment and wages remains intact. Weaker than expected wage growth seen in traditional metrics is thus attributable to compositional effects. These effects should lessen as unusually high levels of underemployment continue to recede (although demographic change will continue as high wage workers retire), and thus traditional wage metrics should accelerate. That is the Fed’s expectation as well.

But what about inflation? Will inflation necessarily move higher as labor markets improve further? That is still an open question. Rising wages would be evidence of decreasing economic slack, and Federal Reserve Chair Janet Yellen has said that she anticipates inflation to rise back to target as slack diminishes and the transitory impacts on inflation wane. Indeed, if the economy reverts to an equilibrium similar to that of the late 2000’s, we would expect both wage growth and inflation to both move roughly 100bp higher as unemployment declines toward 4.5%. In such a scenario, real wage growth would be unchanged. Indeed, adjusting the Atlanta Fed numbers for inflation indicates that real wage growth has already returned the late 2000’s range of 1.5-2% year-over-year: 

It appears that the economy transitioned to lower real wage growth relative to the late 1990’s in response to the productivity slowdown. 

Hence I think the base case of rising wages and prices remains reasonable – assuming sufficient cyclical momentum to carry unemployment lower still. But how much tightening can the economy weather before that cyclical momentum wanes? Therein lies the Fed’s challenge. Employment indicators tend to be lagging, and the economy may already be already easing into a soft patch. Conor Sen sees that the drivers of growth this cycle are abating, and hence activity will need to be a transition to new drivers. Note also signs that the US credit cycle is already tightening and the rising levels of distressed debt (only a third of which is oil and gas related). In other words, if the economy is indeed at an inflection point with credit conditions already tighter, the room for tightening is likely limited – and the room for error higher. This is likely more so the case in a world of low interest rates; in such a world, policy might turn tighter more quickly than in previous cycles.  

Bottom Line: Just how data-dependent is the Fed when it comes to December? Not much, I think. They are likely just looking for evidence that basic labor market trends remain intact to justify pulling the pin on higher rates. Absent any sharp financial disruptions or disastrous data, it looks like we are on the final countdown to the first rate hike of this cycle. Beyond that, they will proceed very cautiously; this is especially the case if they don’t see evidence of still-declining slack in the form of rising wages and inflation. And if the economy turns choppy as the drivers of recent growth loose their momentum, policy will turn choppy as well. Indeed, in such an environment, future rate hikes would likely comes in fits and starts. Thus while 100bp of tightening is a reasonable baseline for next year, the path is not likely to be a smooth 25bp every other meeting. That will likely pose some interesting communications challenges for the Fed.

Mission Accomplished

Federal Reserve policymakers have pretty much taken all of the mystery out of this next meeting. Federal Reserve Vice Chair Stanley Fischer, via Reuters:

"In the relatively near future probably some major central banks will begin gradually moving away from near-zero interest rates," Fed Vice Chairman Stanley Fischer told the San Francisco Fed's biannual Asia Economic Policy conference.

"While we at the Fed continue to scrutinize incoming data, and no final decisions have been made, we have done everything we can to avoid surprising the markets and governments when we move, to the extent that several emerging market (and other) central bankers have, for some time, been telling the Fed to 'just do it'."

New York Federal Reserve President William Dudley, via Reuters:

The Federal Reserve should "soon" be ready to raise interest rates as U.S. central bankers grow confident that low inflation will rebound and that employment remains stable, William Dudley, the influential head of the New York Fed, said on Friday.

"We hope that relatively soon we will become reasonably confident that inflation will return to our 2 percent objective," he said at Hofstra University. Dudley said it was "very logical" to expect that the Fed's inflation and employment conditions would be met "soon," allowing policymakers to "start thinking about raising the short-term interest rates."

Atlanta Federal Reserve President Dennis Lockhart, via CNBC:

A top Federal Reserve official said Thursday he is "comfortable" with raising the federal funds rate "soon," as concerns about low inflation and global risks are not persuasive enough to keep interest rates near zero.

"I'm comfortable with moving off zero soon," said Atlanta Fed President Dennis Lockhart in prepared remarks.

San Francisco Federal Reserve President John Williams, via Reuters:

"The data I think have been overall encouraging, especially on the labor market," San Francisco Fed President John Williams told reporters after a conference at University of California Berkeley's Clausen Center.

"Assuming that we continue to get good data on the economy, continue to get signs that we are moving closer to achieving our goals and gaining confidence getting back to 2-percent inflation... If that continues to happen there's a strong case to be made in December to raise rates."

Obviously serial dissenter Richmond Federal Reserve President Jeffrey Lacker is also looking for a rate hike. And so too is Cleveland Federal Reserve President Loretta Mester. To be sure, they all give a nod to “data dependence,” implying that a rate hike is not a sure thing. But, barring an outright collapse in financial markets, it is very difficult to see the data evolve between now and December 15-16 in such a way that the Fed suddenly has a change of heart. And note there is little reason for them to think at this point that growth has slowed well below trend. It is widely expected that Q3 GDP is this week revised up to 2.1% while current quarter GDP is tracking at 2.3%. While in 1990s terms these are not staggering numbers, in 2010 terms they exceed the Fed’s estimate of potential GDP growth. And with more and more Fed officials convinced the economy is operating near full employment, anything over 2% raises worries on Constitution Avenue that the economy might overheat.

Now, we still have one employment report ahead of us. Aside from the now-reversed equity declines in August, recall from the last minutes that uncertainty regarding the labor market helped stay the Fed’s hand:

In assessing whether economic conditions and the medium-term economic outlook warranted beginning the process of policy normalization at this meeting, members noted a variety of indicators, including some weaker-than-expected readings on measures of labor market conditions, and almost all members agreed it was appropriate to wait for additional information to clarify whether the recent deceleration in the pace of progress in the labor market was transitory or reflected more persistent factors that might jeopardize further progress.

It would seem that the October labor report put an end to those concerns. Consequently, the following comes into play:

Members emphasized that this change was intended to convey the sense that, while no decision had been made, it may well become appropriate to initiate the normalization process at the next meeting, provided that unanticipated shocks do not adversely affect the economic outlook and that incoming data support the expectation that labor market conditions will continue to improve and that inflation will return to the Committee's 2 percent objective over the medium term.

I suspect that only an outright disaster in the November labor report would prompt the Fed to take a pass at the December meeting. It is just simply the case that given their Phillips curve framework, they are running out of reasons not to raise rates. They would need enough weak data to fundamentally alter their outlook to the downside, and it is hard to see that happening in the short time remaining.

Consequently, it is hard to come to any other conclusion than that they are going to raise the target range on the federal funds rate in December. In Fedspeak, they might as well be screaming it into your ears.

While they may be taking the mystery out of the first rate hike, however, they are trying to put the mystery into subsequent rate hikes. Lockhart, via Reuters:

"The pace of increases may be somewhat slow and possibly more halting than historic episodes of rising rates," Lockhart said in a speech to the DeKalb Chamber of Commerce in Atlanta.

Williams, via Reuters:

"We definitely do not want to, either through our actions or our words, indicate a preference for a very mechanical path of interest rates, whether it’s every other meeting or however you think about it," Williams said. "Since economic data can surprise on the upside and the downside, maybe there will be opportunities to show we are data dependent."

And St. Louis Federal Reserve President James Bullard, via Bloomberg:

“When we had a normalization in 2004 to 2006 we moved at the same 25 basis points per meeting for 17 meetings in a row,” Bullard said. “I am virtually certain that was not optimal monetary policy. That was a very mechanical approach to increasing rates. This time I am hopeful we can be more flexible and reactive to data.”

How will they communicate uncertainty in the path of rate hikes? I wonder if they can simply retain this sentence in the next statement:

In determining whether it will be appropriate to raise the target range at its next meeting, the Committee will assess progress--both realized and expected--toward its objectives of maximum employment and 2 percent inflation.

It seems like this could be used to convey uncertainty in subsequent meetings, especially if they choose not to hike in January.

Bottom Line:  The Fed is set to declare “Mission Accomplished” at the next FOMC meeting. Indeed, many policymakers have already said as much. Absent a very significant change in the outlook, failure to hike rates in December would renew the barrage of criticism regarding their communications strategy that prompted them to highlight the December meeting in their last statement. Once they have communicated their intentions for subsequent rate hikes, they will turn their attention to the issue of normalizing the balance sheet. Even though officials have not committed to a specific path, I am working with a baseline of 100bp of tightening between now and next December, or roughly 25bp every other meeting. I expect that by the second quarter of next year they will begin communicating the fate of the balance sheet. Whether they should hike or not remains a separate issue. Over the next twelve months we will learn the extent of which the Federal Reserve can resist the global downward pull of interest rates. Other central banks have been less-than-successful in their efforts to pull off of the zero bound – not exactly a hopeful precedent.

Onto The Next Question

It would seem that a December rate hike is all but certain barring some dramatic deterioration in financial conditions. The October employment report should remove any residual concerns among FOMC members over the underlaying pace of activity, clearing the way for the Fed to make good on the strongly worded October FOMC statement. Given the resilience of recent trends, it is tough to see that even a weak-ish November employment report would dissuade the Fed from hiking rates. Quite frankly, regardless of whether you think they should hike rates, if they don't hike rates, the divergence between what they say and what they do would become truly untenable from a communications perspective.

Nonfarm payrolls jumped 271k in October, a relief after two weaker reports. Note though that the three-month moving average still indicates that job growth has lost some momentum:

That said, momentum remains sufficient to sustain ongoing improvement in a wide array of labor market indicators. Those pervasively identified by Federal Reserve Chair Janet Yellen:

Notably, wage growth accelerated, giving fresh hope that it has broken out of its multiyear doldrums. The Fed will see this as evidence that their estimates of the natural rate of unemployment are more right than wrong:

Another way to see that wage growth may be set to break higher:

If nominal wage growth were to break higher, would that reflect the impact of productivity gains, margin compression, or higher inflation? Your view on that question will influence your rate outlook. 

With unemployment edging below the Fed's current estimate of the natural rate (note that we get updated forecasts in December) and wages showing signs of life, it seems that the Fed is positioned to move forward with a rate hike in December.  This is especially if they want to make good on their promise to hike rates at a gradual pace.  San Francisco Federal Reserve President John Williams reiterated that point last week. Via the Wall Street Journal:

“An earlier start to raising rates would also allow a smoother, more gradual process of policy normalization, giving us space to fine-tune our responses to any surprise changes in economic conditions,” Mr. Williams said. “If we were to wait too long to raise rates, the need to play catch-up wouldn’t leave much room for maneuver,” he said.

Note that the first hike and pace of tightening were never really separate as the Fed would like you to believe. Williams makes clear the the pace was in fact dependent on the timing of the first hike. The earlier they start, the more gradual the subsequent pace. 

The question now arises, however, of what is "gradual"? The general consensus is the "gradual" means 25bp every other meeting. St. Louis Federal Reserve President James Bullard says there is not fixed definition as of yet. Via Reuters:

"Once 'liftoff' occurs the debate will immediately shift to when is the next move going to come? How fast is the pace of increases going to be? ... What does 'gradually' actually mean?" Bullard said. "That is going to be a hot debate and we won't really have credibility as a committee for the notion of gradualness until we make that second move."

Has the Fed already waited too long to sustain a path of 25bp every other meeting? That is what we should be asking. Indeed, I believe the next labor report will have more implications for the January meeting than the December meeting. Anxiety among Fed officials regarding whether or not they are falling behind the curve is inversely proportional to the unemployment rate. If it ticks down to 4.8% in the November report, they will start to get very nervous that 25bp every other meeting is not tenable. It of course goes without saying that if core-inflation starts to firm in the next two months and tend toward trend more quickly than anticipated, policymakers will break into a cold sweat.

Still unknown is how rate hikes will interact in the global environment. Fed Governor Lael Brainard has yet to give up her concerns. Via MarketWatch:

Brainard said the "feedback loop" between market expectations of divergence between the U.S. and its major trading partners and financial tightening in the U.S. means that "material restraint to U.S. conditions is already in place."

How much tightening Fed tightening can the US sustain in a world driven the zero lower bound globally? Such concerns are generally downplayed by Fed officials; that lack of concern is something I view as a key risk. The tipping point between loose and tight financial conditions is likely lower than in the past. The Fed may blow past that tipping regardless of how fast they hike rates. In some sense, one can argue that the end point for rate hikes is more important than whether the Fed moves on average at 12.5bp or 25bp every meeting.

Bottom Line: The debate is shifting. It is soon to be no longer about the first rate hike. Fed officials, the question is shifting from whether they should go at all to whether they waited too long.  

What 2016 Might Bring

I recently predicted the following:

One of two things is going to happen. Either the US economy is or will soon be slowing on the back of already tighter financial conditions. Or the US economy will soon be slowing on the back of future tighter financial conditions as directed by the Federal Reserve.

My baseline expectations for next year need more explanation, particularly in light of the weak third quarter GDP report and the early signals on fourth quarter growth via the Atlanta Fed’s GDPNow tracker (currently at the low-end of consensus). Three caveats, however, to keep in mind. First, I avoid over-analyzing the quarterly fluctuations in GDP preferring instead to track trends over a longer period. Second, similarly, the initial release will be subject to substantial revision. Third, the Atlanta Fed number may or may not evolve over the course of the quarter; where it is now is not necessarily where it will be when fourth quarter data is released.

That said, GDP growth slowed noticeably in the third quarter, dragging down recent trends: 

Negative inventory adjustment, however, was a significant factor. When we look at recent trends in final sales to domestic purchases, domestic momentum remains solid:

Generally, housing, autos, services, and the government sectors remain solid. The soft spots are the external sector and manufacturing. These two are obviously related; weakness in manufacturing is closely tied to a stronger dollar and reduced activity in the oil and gas exploration. ISM surveys reveal a striking divergence between the manufacturing and services sides of the economy:

It is thus quite arguable that, after accounting for inventories, little momentum has been lost. The softening of the job growth, however, suggests that the underlying pace of growth has pulled back from full throttle (at least our current definition of full throttle):

Perhaps then growth has in fact softened, possibly a consequence of already tighter monetary policy. Minneapolis Federal Reserve President Narayana Kocherlakota:

In mid-2013, the FOMC announced its intention to taper its ongoing asset purchase program. We can see that this announcement represented a dramatic change in policy from the sharp upward movements in long-term bond yields that it engendered. Personally, I interpret this policy change back in 2013 as the onset of what the Committee currently intends to be a long, gradual tightening cycle. As I noted earlier, we would typically expect that such a change in monetary policy should affect the economy with a lag of about 18 to 24 months. Viewed through this lens, the slow rate of labor market improvement in 2015 is not all that surprising.

We will get a reading on the labor market Friday to help confirm or deny recent trends. Suppose the numbers both this month and next are better than expected, thus belying the recent softness. What will be the Fed’s reaction? I think it is fairly safe to say the “raise rates” contingent will have the upper hand in December, thus formally beginning the “normalization” process with a first rate hike of the cycle.

In other words, if growth is not in fact slowing, then the Federal Reserve will likely soon take action to slow growth. How many rate hikes follow? And how rapidly do they follow? The Fed appears to believe that they have roughly 375bp ahead of them and can raise rates every other meeting to get there. What actually happens will depend on how hard they think they will be running up against any constraints in the economy. As a summary indicator, note that the unemployment rate already sits at something near policymaker’s estimate of the natural rate of unemployment:

My interpretation of the Fed’s intentions is that they would like to see the unemployment rate temporarily stabilize at something below the natural rate to allow for further reduction in underemployment. To accomplish this job growth will need to slow over the next year to that necessary to absorb growth in the labor force. What does that mean for the numbers? San Francisco Federal Reserve President John Williams offers what is probably a reasonable middle ground among officials:

As we make our way back to an economy that’s at full health, it’s important to consider what constitutes a realistic view of the way things will look. The pace of employment growth, as well as the decline in the unemployment rate, has slowed a bit recently…but that’s to be expected. When unemployment was at its 10 percent peak during the height of the Great Recession, and as it struggled to come down during the recovery, we needed rapid declines to get the economy back on track. Now that we’re getting closer, the pace must start slowing to more normal levels. Looking to the future, we’re going to need at most 100,000 new jobs each month. In the mindset of the recovery, that sounds like nothing; but in the context of a healthy economy, it’s what’s needed for stable growth. 

As the next year unfolds, what we want to see is an economy that’s growing at a steady pace of around 2 percent. If jobs and growth kept the same pace as last year, we would seriously overshoot our mark. I want to see continued improvement, but it’s not surprising, and it’s actually desirable, that the pace is slowing.

All else equal, if they are not seeing evidence of that slowing by the middle of next year I would expect them to accelerate the pace of rate increases. That is probably when we need to worry about overshooting. Not so much from the faster rate increases, but from the failure to account for policy lags. It may be a challenge to see the impacts of policy tightening early on if rate hikes are at a glacial pace. Hence the Fed may erroneously believe they need to play “catch-up” more than is truly necessary.

Overshooting, however, is a consideration for a later day. At this point it is sufficient to recognize that, at least under the current monetary policy framework, either the economy will slow by itself or the Fed will eventually work to force it to slow. That would seem to suggest that growth is at or past its peak for this cycle. That is the situation I am most wary of at the moment, leading me to the conclusion that growth is headed down in 2016.

I am not wedded to that scenario. I can envision sustained higher growth on the back of either faster than anticipated labor force growth or faster productivity growth. Recent trends tend not to be terribly supportive, but nonetheless I remain watchful that those trends shift. Indeed, perhaps we will see productivity rise as firms react to tighter labor markets. Such a scenario could deliver a sustained growth with accelerating wages. That would obviously be something of a win-win situation.

To be sure, the inflation outlook has an impact on the Fed's timing, but it remains something of a wildcard. The Fed expects to normalize only after they are reasonably confident that inflation will return to target. Two more solid job reports are enough to get there. The pace of subsequent rate hikes depends on the evolution of inflation relative to that target. As Federal Reserve Chair Janet Yellen said today, via the Wall Street Journal:

Referring to recent remarks by Fed governor Lael Brainard on the subdued state of U.S. inflation, Ms. Yellen told lawmakers that “if we were to move, say in December, it would be based on an expectation -- which I believe is justified -- that with an improving labor market and transitory factors fading, that inflation will move up to 2%. But of course if we were to move, we would need to verify over time that expectation was being realized, and if not, adjust policy appropriately.”

Near term inflation perked up a bit in September, but still remains below target:

One might think that persistently low inflation eventually wears on inflation expectations. Yellen raised this concern in September:

Although the evidence, on balance, suggests that inflation expectations are well anchored at present, policymakers would be unwise to take this situation for granted. Anchored inflation expectations were not won easily or quickly: Experience suggests that it takes many years of carefully conducted monetary policy to alter what households and firms perceive to be inflation's "normal" behavior, and, furthermore, that a persistent failure to keep inflation under control--by letting it drift either too high or too low for too long--could cause expectations to once again become unmoored. Given that inflation has been running below the FOMC's objective for several years now, such concerns reinforce the appropriateness of the Federal Reserve's current monetary policy, which remains highly accommodative by historical standards and is directed toward helping return inflation to 2 percent over the medium term.

Interestingly, the University of Michigan’s survey of inflation longer-term inflation expectations continues to drift lower just as the Fed is considering rate hikes:

Contrast with the cycle of tightening in the middle of the last decade:

The accuracy of survey-based measures is in doubt, however. For example, via the St. Louis Federal Reserve, economist Kevin Kliesen concludes:

Going forward, most Federal Reserve officials expect inflation to eventually return to 2 percent. But when using measures of inflation expectations to forecast future inflation, policymakers and forecasters should focus on market-based measures of inflation expectations. They are much more accurate than survey-based measures.

Yellen, however, hesitates to embrace market-based measures of inflation expectations (although I suspect she would quickly embrace them if they headed higher). Discarding both measures thus leaves us with little guidance, unfortunately. My take is that there is probably some information from the direction of both measures, and that information is generally not supportive of the Fed’s confidence that inflation will return to target in a timely fashion. The Fed would have a hard time justifying ongoing hikes, even if the economy outperforms their expectations, if inflation remains tame. My suspicion is that under such a scenario the Fed would pivot away from their current inflation framework to financial stability concerns to justify tighter policy. 

Bottom Line: I tend to believe that growth has peaked for this cycle, or, more accurately, that sustaining these growth rates will likely require faster productivity or labor force growth. Indeed, it appears the Fed will force such an outcome if they remain committed to their basic policy framework. This seems like a reasonable baseline from which to think about the next 4 or 5 quarters. Productivity growth could pick up such that a stabilizing unemployment rate remains consistent with steady growth. Assuming growth is not yet softening, a 25bp rate hike every other meeting beginning in December is also a reasonable baseline for monetary policy; if the Fed doesn't see that having an impact, they will likely step up the pace. It should go without saying that a slowing economy is not to be equated with a recession.