Tim Duy's Fed Watch

December Looking Good. But...

FOMC doves squeezed out another victory at last week’s meeting. But can they do it again in December?

As was widely expected, the Fed held rates steady at the September FOMC meeting. That said, the meeting was clearly divisive, with three dissents, all from regional bank presidents. And the accompanying statement leaned in a hawkish direction – the committee noted that near-term risks were “balanced” and that the case for a rate hike had “strengthened.” Moreover, only three of the participants did not expect a rate hike before year end.

And if that was not enough, during her press conference, Federal Reserve Chair Janet Yellen suggested the bar to a December rate hike was low:

…most participants do expect that one increase in the federal funds rate will be appropriate this year and I would expect to see that if we continue on the current course of labor market improvement and there are no major new risks that develop and we simply stay on the current course.

Sounds like December is a go. But markets are not entirely convinced, with participants pricing in a roughly 60% chance of a rate hike. Perhaps this pricing reflects post-election economic risk. Or perhaps it reflects the possibility that the doves can stare down the hawks one more time before the composition of the Board changes next year.

Can they? That question requires understanding what happened to squash the parade of Fed presidents looking for a rate hike in September. What happened were Federal Reserve Governors Lael Brainard and Daniel Tarrullo. Brainard in particular laid down the intellectual framework ahead of the FOMC meeting, arguing that the potential for further labor market improvement and asymmetric policy risks justified a steady hand at this meeting. Yellen and the rest of the Board bought into this story. The hawks could squawk all they wanted, but the votes just weren’t going to go in their favor.

This episode provided two important lessons. The first is that if you haven’t been taking Brainard seriously this past year – ever since her bombshell speech last October – you have been doing it wrong. The second is that a small group of governors can have a much larger influence on policy than a large group of presidents. There are lots of presidents, and they talk a lot, so their message is louder. But the power rests in the Board.

Indeed, this asymmetry of power is why the relative lack of speeches from Board members is one of the Fed’s biggest communication failures. The people driving policy shouldn’t be waiting until the Friday before the blackout period to begin delivering their message.

Now consider the dots. There remain three “no hike” dots for 2016. I think it is reasonable to believe those three dots belong to Tarullo, Brainard, and Chicago Federal Reserve President Charles Evans. If true, that suggests that Tarullo and Brainard are at the present time considering making another dovish stand at the December meeting. To do so, they need to keep Yellen on their side.

During the press conference, Yellen revealed that she remains attached to a preemptive view of policy. Since monetary policy operates with long lags, it is important that policy responds to inflationary threats before they emerge. She also rejected a “whites of their eyes” approach to policy, or the suggestion by Evans that they Fed waits until core inflation hits two percent before they hike rates. These concerns are balanced against Brainard’s argument that they can’t be sure they have yet achieved full employment.

Hence, and as I said ahead of the meeting, I think that if unemployment dips between now and December, or progress on underemployment resumes, or inflation moves closer to target, the hawks will win as Yellen’s support will shift toward a rate hike. And these things can all be reasonably expected given the current course of job growth, which is in excess of the Fed’s estimate of what is necessary to absorb labor force growth. For the doves to have a decent chance of holding back the hawks one more time, progress on these points needs to remain stalled.

Regardless of a December hike or not, the Fed continues to mark down the expected path of policy. The median projected Fed funds rate dropped 50bp for both 2017 and 2018, continuing the pattern of the Fed moving toward the market rather than vice-versa. And note that the changing composition of the FOMC next year will allow for this dovish message to come through. This meeting’s dissenters will all be replaced with presidents that are on average more dovish. Consider this ordering of monetary policy makers via Julia Coronado of Graham Capital, modified to show the shift of voters for next year:

Voting presidents will be more aligned with the preferences of the governors. This should help ease some of the recent communications challenges even if the governors maintain their relative silence.

Bottom Line: Doves on the Board continue to delay the preemptive strike on inflation. Stalling gains on unemployment and underemployment gave them the ammunition to stand their ground. If those gains resume, doves will fall prey to the hawks at the next meeting. But they will have an easier time maintaining a shallow path of policy next year, and hopefully are better set to communicate that path.

Ahead Of The FOMC Meeting

A roundup of Fed-related stories and viewpoints ahead of the FOMC meeting. First, Jeanna Smialek at Bloomberg sees danger lurking in the new dot plot:

Janet Yellen will frame a decision this week to forgo an interest-rate increase as necessary to achieve the Federal Reserve’s economic goals. Donald Trump and his supporters are likely to frame it as political.

That’s because the central bank on Wednesday will also release fresh “dot plot” projections which will probably show policy makers see one quarter-point rate hike by the end of the year. Such a forecast would be widely interpreted as a sign that a hike is coming at the Fed’s December meeting, instead of at the November gathering, which comes a week before the U.S. presidential election and isn’t accompanied by one of the chair’s quarterly press conferences.

Problem is, having the dot plot signal a December move comes with political baggage...

The political baggage is the timing of the rate hike around a presidential election. Why wait on a rate hike now only to signal that one is coming in December? Detractors will claim that the Fed doesn't want to derail the economy and with it the Democrat's hope of retaining the White House. This despite, as Joe Gagnon notes in the article, politics has little if any impact on the rate setting decision. But this isn't about reality, it is about perception. And, politically, the optics just aren't great.

It seems to me that the Fed is taking a political hit on top of what is likely to be the communications hit if, as is reasonably assumed, the dot plot signals a quarter-point hike in December. That would be a pretty strong calendar-based signal of their intentions. Given there is only a few months left in the year, they have to be pretty confident in the outlook to send such a signal. Which raises the question that if you were so confident, why not hike rates now? And if you send such a strong signal now, is that lowering the bar on the kind of data you need to support a hike? And then are you hiking because of perceived past commitment, a need to maintain "credibility," rather than the data? But doesn't that make the Fed more susceptible to policy errors?

In my opinion, the dots outlived their usefulness when they signaled a pace of policy tightening that never happened. They were a great tool for credibly committing to zero for a long period. But that very credibility made them a terrible tool when the time came for tighter policy. They were perceived as a promise because such perception followed logically from the previous promise of low rates. Now they just appear as a series of broken promises. Worse yet, the Fed might feel tied to those promises when they shouldn't be. 

The Fed really needs to rethink the dot plot. Use it as a tool when it can be most effective; pull it when it detracts from the message.

Meanwhile, former Minneapolis Federal Reserve President Narayana Kocherlakota, writing at Bloomberg View, says the Fed is about to make a mistake regardless of what they do:

More than seven years after the recovery began in mid-2009, inflation remains below the central bank's 2-percent target...Worse, markets appear to be losing confidence that the Fed will ever reach its target: Yields on Treasury bonds suggest that traders expect inflation to average less than 2 percent five to 10 years from now. As the experience of the Bank of Japan indicates, restoring such confidence is not easy...The Fed is also falling short of its goal of "maximum" employment. 

Kocherlakota concludes that the Fed should be easing policy, so holding and raising rates are both mistakes at this juncture. 

But one does not have to go far for an opposing view. The editorial board at Bloomberg has a different idea:

The best it can do is press cautiously ahead on normalizing monetary policy, explain what “normal” now means, and promise to keep an open mind as new information comes in. What this requires right now, it should also say, is a quarter-point rise in interest rates.

The editorial board dismisses Kocherlakota as missing the bigger picture:

What this kind of analysis leaves out is the growing threat to future financial stability. Very low interest rates (together with a massively enlarged central-bank balance sheet, courtesy of quantitative easing) have supported demand as intended, albeit with ever-diminishing effectiveness; at the same time, however, they’ve artificially boosted financial-asset prices and distorted normal patterns of risk-taking in financial markets.

Because interest rate are low, they must be "artificially" low and thus distorting something in the economy. The insinuation is that the Fed can simply raise interest rates and the economy will jump back into a happier equilibrium with no distortions and no negative impact. Good luck with that.

If interest rates were truly too low, then their should be much more economic activity and upward pressure on inflation than currently exists. Whatever distortions currently exist must not be exerting a broad impact and thus are fairly small; monetary policy is a blunt tool to use on small distortions. Nor is it evident that even a fairly large rate hike would stop an asset bubble - at least not without a cost. San Francisco Fed researchers concluded:

What is the takeaway then? Slowing down a boom in house prices is likely to require a considerable increase in interest rates, probably by an amount that would be widely at odds with the dual mandate of full employment and price stability. Moreover, the Fed would need a crystal ball to foretell house price booms. In restraining asset prices, while the power of interest rate policy is uncontestable, its wisdom is debatable.

So hiking rates now to try to stop a bubble will likely end in lower rates later. In other words, to use rate policy to try to calm financial markets, you better be very, very sure you are actually facing a widespread threat to the economy. And I don't see anything that justifies that level of certainty. The Financial Crisis was the last war; policymakers need to be wary about always fighting the last war.

Not everyone believes the Fed will hold steady tomorrow. Via Bloomberg:

Two of the Fed’s 23 preferred bond-trading partners -- Barclays Plc and BNP Paribas SA -- are betting against their peers and the bond market by forecasting officials will raise rates Wednesday. It’s the first time more than one dealer has gone against the consensus during the week of a policy meeting since last September, data compiled by Bloomberg show. Economists at both banks say traders have too steeply discounted officials’ intent to hike after the Fed has remained on hold for longer than expected.

I think this is highly unlikely. There are some heavy hitters pushing to holding rates steady. I would not underestimate the power of a few dovish board members, especially if they don't want to roll over on a rate hike like last December. Moreover, the Fed doesn't like to surprise market participants. They don't need 100% certainty, but they need something better than the current odds hovering between 10 and 20%. The hawks know this, and don't like the outcome of the meeting being a foregone conclusion. That said, if the Fed does hike, the handful of analysts who called for a rate hike will look brilliant. And they should get the credit where credit is due in that circumstance. 

And for my views on the meeting, see my piece in Bloomberg this week. 

Is Pushing Unemployment Lower A Risky Strategy?

The unemployment is closing in on the Fed's estimate of the natural rate of unemployment:
Consequently, Fed hawks are pushing for a rate hike sooner than later in an effort to prevent the economy from "overhearing." This overheating is argued to set the stage for the next recession. For instance, see San Francisco Federal Reserve President John Williams:

History teaches us that an economy that runs too hot for too long can generate imbalances, potentially leading to excessive inflation, asset market bubbles, and ultimately economic correction and recession. A gradual process of raising rates reduces the risks of such an outcome. It also allows a smoother, more calibrated process of normalization that gives us space to adjust our responses to any surprise changes in economic conditions. If we wait too long to remove monetary accommodation, we hazard allowing imbalances to grow, requiring us to play catch-up, and not leaving much room to maneuver. Not to mention, a sudden reversal of policy could be disruptive and slow the economy in unintended ways.

In his Bloomberg View column, former Minneapolis Federal Reserve President Narayana Kocherlakota questions whether there is much theory behind this contention:

Some Fed officials worry that “overheating” could trigger a recession. (I don’t understand the precise economic mechanism, but let’s leave that aside.)

Kocherlakota was specifically referring to the risks of undershooting the natural rate of unemployment. New York Federal Reserve President William Dudley summarized his perception of that risk in January of this year:

A particular risk of late and fast is that the unemployment rate could significantly undershoot the level consistent with price stability. If this occurred, then inflation would likely rise above our objective. At that point, history shows it is very difficult to push the unemployment rate back up just a little bit in order to contain inflation pressures. Looking at the post-war period, whenever the unemployment rate has increased by more than 0.3 to 0.4 percentage points, the economy has always ended up in a full-blown recession with the unemployment rate rising by at least 1.9 percentage points. This is an outcome to avoid, especially given that in an economic downturn the last to be hired are often the first to be fired. The goal is the maximum sustainable level of employment—in other words, the most job opportunities for the most people over the long run.

I don't know that there is an economic mechanism at work here. I don't know that there is a law of economics where the unemployment can never be nudged up a few fractions of a percentage point. But I do think there is a policy mechanism at play. During the mature and late phase of the business cycle, the Fed tends to overemphasize the importance of lagging data such as inflation and wages and discount the lags in their own policy process. Essentially, the Fed ignores the warning signs of recession, ultimately over tightening time and time again.

For instance, an inverted yield curve traditionally indicates substantially tight monetary conditions. Yet even after the yield curve inverted at the end of January 2000, the Fed continued tightening through May of that year, adding an additional 100bp to the fed funds rate. The yield curve began to invert in January of 2006; the Fed added another 100bp of tightening in the first half of that year.

This isn't an economic mechanism at work. This is a policy error at work.

Kocherlakota offers another important point:

It's easy to imagine, though, that many people would be willing to trade the risk of recessionary pain in 2019 and 2020 for the near-term gain of 2017 and 2018. They might even believe there's some chance that policy 2 will generate an outstanding outcome -- if, for example, the long-run unemployment rate is actually lower than the Fed thinks it is.

The Fed seems to place almost zero weight on the probability that the natural rate of unemployment is significantly below their estimates. In their view, only bad things happen when the unemployment rate drifts much below 5%.  

Bottom Line: The Fed thinks the costs of undershooting their estimate of the natural rate of unemployment outweigh the benefits. I am skeptical they are doing the calculus right on this one. I would be more convinced they had it right if I sensed that placed greater weight on the possibility that they are too pessimistic about the natural rate. I would be more convinced if they were already at their inflation target. And I would be more convinced if their analysis of why tightening cycles end in recessions was a bit more introspective. Was it destiny or repeated policy error? But none of these things seem to be true.

Rate Hike Hopes Fading Fast

The next FOMC meeting is just two weeks away. Fed hawks had hoped that this was their moment in the sun. I suspect they will need to wait another three months before their next opportunity to act. Signs of a second half rebound are likely too tentative for the doves to tolerate a rate hike. I don't think they will roll over as easily as they did last December.

The August employment report was not terrible. Not by any measure. On the positive side, labor supply is reacting to both demographic changes and stronger demand:

The demographic shift - essentially, the aging of the Millennials toward their prime age working years - is I believe a powerful secular force supporting the economy. That said, the Fed needs to ensure cyclical forces do not undermine the economy. And that is where the story becomes tricky. Is the economy slowing sufficiently on its own that the Fed should refrain from rate hikes? Or is the slowing still insufficient to quell the inflationary pressures Fed hawks in particular believe to be building?

On first take, the slowing in payroll growth is modest:

And arguably sufficient to place additional downward pressure on the unemployment rate. Cleveland Federal Reserve President Loretta Mester recently repeated this view, which is widely held within the FOMC. Via Reuters:

Mester, a voting member on the Fed's policy-setting committee, had earlier in the day told a philanthropy conference that the U.S. economy probably needs to generate between 75,000 and 150,000 jobs per month to keep the jobless rate stable.

Hiring has been stronger than that this year and the U.S. jobless rate is currently at 4.9 percent.

"The economy is basically at full employment," Mester said.

This "full employment" view is also evident in the Fed's estimate of the natural rate of unemployment:

This, not inflation directly, seems to be driving Fed hawks toward a rate hike. See former Federal Reserve President Narayana Kocherlakota here. It is the perceived threat of inflation, not the actual, realized threat of inflation. 

Fed hawks will also point toward wage growth as evidence of tighter labor markets that foreshadows inflationary pressures:

Fed doves, however, will not be without their own interpretation of the data. The flattening of the unemployment rate could indicate supply side pessimism on the part of the hawks. That is the positive story that still fits with a no hike scenario. A more negative story is that the flat unemployment rate is consistent with late cycle patterns:

 

Similarly, progress toward reducing underemployment has stalled noticeably, leaving underemployment at very high levels:

Perhaps the household data is picking up a degree of slowing not yet evident in the establishment data? And on the establishment side, temporary help services payrolls are holding in a late cycle pattern as well:

As far as wages are concerned, Fed doves will say that wage growth is still anemic in comparison with past cycles and - they should add - that wages are a lagging indicator. The Fed should be paying much more attention to forward indicators. And those forward indicators remain tentative at best. The hawks' basic case is not just that the economy is at full employment, but that a second half rebound will send it beyond full employment. And while consumer spending supports the second half rebound story:

the ISM reports draw that into question. Today's service sector report was particularly disconcerting with weakness across the board - the sharp drop in new orders should give FOMC members reason for caution. Doves will thus say the Fed can't count their chickens before they hatch. And this is especially important given that the Fed continues to miss its inflation target, and a misstep at this juncture with overly tight policy will basically guarantee they miss it for the next five years as well.

Indeed, while Fed hawks such as Vice Chair Stanley Fischer and Boston Federal Reserve President Eric Rosengren see progress toward the inflation goals, Peter Olson and David Wessel, writing in the WSJ, conclude:

The inflation rate is higher now than it was in 2015. But over the course of 2016 we’ve seen no apparent progress toward the 2% inflation target. If anything, the inflation rate in January was closer to the Fed’s goal than in July. So it’s increasingly difficult for Fed officials to rely on current inflation numbers as a justification for raising rates. Higher inflation might be just around the corner, but we haven’t seen it yet.

I agree. The "progress" that Fischer and Rosengren point to occurred early in the year, mostly in January. Recent trends have been less promising.  The hawks "inflation is here" story is not particularly compelling. Indeed, I would say it borders on disingenuous. Moreover, I suspect the inflation numbers will prompt strong opposition to a rate hike this month. Recall from the recent minutes:

 A couple of members preferred also to wait for more evidence that inflation would rise to 2 percent on a sustained basis.

I suspect these two members were Governors Lael Brainard and Daniel Tarrullo. My guess is that neither will roll over on a rate hike as they did last December; I think they probably question the wisdom of the outcome of that meeting. Furthermore, I think they pull Governor Powell and ultimately New York Fed President William Dudley to their side. St. Louis Federal Reserve President James Bullard is ambivalent about when the next 25bp hike occurs; in his framework, the Fed is already within spitting distance of the correct policy stance. He won't push for a hike. And I suspect that Chair Janet Yellen will thus ultimately see too little consensus to support a rate hike. 

Bottom Line: Despite being near the consensus view of full employment, incoming data on the second half remains too tentative to support a rate hike this month. This is especially the case given lost momentum in the labor market, particularly with regards to underemployment, and the weak inflation numbers. Hence I do not anticipate a rate hike in September. Why might I be wrong? Aside from just being wrong on the Fed's likely interpretation of the incoming data, perhaps because I have underestimated the Fed's perception that the risks are not really asymmetric - that they have all the tools they need to fight the next recession even if they are at the zero bound - or that the Fed views financial stability concerns as trumping the inflation outlook. 

Thoughts Ahead Of The Employment Report

The August employment report has come to be seen as the deciding factor in the Fed's upcoming decision on rates. See Sam Fleming at the Financial Times here. Maybe this is the case, maybe not. I hope not. Hinging policy on the first print of nonfarm payrolls - a volatile, heavily revised number - would be pretty low quality policy making. 

I keep coming back to this by Federal Reserve Chair Janet Yellen from back in December:

...total real private domestic final purchases (PDFP)--which includes household spending, business fixed investment, and residential investment, and currently represents about 85 percent of aggregate spending--has increased at an annual rate of 3 percent this year, significantly faster than real GDP. Household spending growth has been particularly solid in 2015, with purchases of new motor vehicles especially strong.  

This was Yellen's way of justifying a rate hike last December in spite of faltering GDP numbers. Trouble is that PDFP continued a downward slide since then:

Final sales here are off roughly 1.5 percentage points from their cycle highs. That is a nontrivial swing. It is no wonder that job growth accelerated in 2013-14 and then decelerated in 2015:

I tend to think there is room for some further deceleration. Note too that progress on reducing underemployment slowed markedly:

and the unemployment rate is flattening out:

Now, you might say that the Fed needs to hike because wages are rising. But I would say that wages are a lagging indicator

and are likely to continue rising even after a recession begins. Overly shifting the policy focus to wage growth would be a red flag in my opinion. I think the Fed tends to focus too much on lagging indicators in the later stages of a business cycle while ignoring their own policy lags. The end result is overly tight policy.

So when I look at the data, I don't see that the August employment report should be a critical factor in a rate hike decision. I think the critical factors should be the Fed's confidence that growth is set to rebound in the second half of the year and the balance of policy risks.

On the first point, while early signals on growth are positive - see the Atlanta Fed GDPNow measure, for example - they are still just early signals. And today's ISM release doesn't indicate that a manufacturing rebound is right around the corner, so maybe that rebound in investment spending just might take more time as well. And auto sales look to have peaked and are flattening out, so that is not likely to be a source of growth and might be slight drag. So, overall, I don't think we have enough data to be confident that growth will rebound just yet.

Regarding the balance of policy risks, that asymmetry has not magically gone away. The Fed has less room to ease than tighten. And inflation remains mired below target:

So I don't see that that the basic calculus here has changed. If the Fed errors by being too loose now, they have plenty of wiggle room on inflation and policy to respond. If they error on by being too tight, they don't have much policy room and they risk holding inflation below 2 percent for another decade. What's that going to do for inflation expectations?

All that said, there appears to be a movement among FOMC members to minimize the asymmetry of the policy risks. First you have Federal Reserve Vice Chair Stanley Fischer arguing that inflation is close enough to target that it shouldn't be a concern. Via Greg Robb at MarketWatch:

And the core measure of the personal consumption expenditure index — the Fed’s favorite measure of inflation — at 1.6% “is within hailing distance” of the central bank’s 2% target, Fischer added.

I am starting to think Fischer is still living in the 1970s. But perhaps more disconcerting is Yellen's final line from her Jackson Hole speech:

But even if average interest rates remain lower than in the past, I believe that monetary policy will, under most conditions, be able to respond effectively.

She is playing down the asymmetric policy risk issue here. Given the experience of the past decade, she is way too complacent in my opinion. And her complacency hinges on the assumption that they now know they (nominal) natural rate of interest is 3 percent. But that has been a moving target. And I don't think that is the signal being sent by the long end of yield curve. 

Then there is the financial stability argument. All I will say on that is the Fed had better be damn certain that they are facing a real risk to the economy before they pull the trigger on that argument. And I don't see how they can be that certain.

Bottom Line: Regardless of the outcome of the employment report, good or bad, I don't see good case for moving next this month. Too many questions about the forecast, and they still face persistently low inflation and asymmetric policy risks. But all that said, there seems to be a large swath of voting members ready to get behind a rate hike. I think the low odds on a rate hike in September is the market's way of telling the Fed that if they do hike, it would be a mistake.