Raphael Bostic
President and Chief Executive Officer
Federal Reserve Bank of Atlanta

Global Interdependence Center: Central Bank Series
Madrid, Spain

November 15, 2018
  • On November 15, Atlanta Fed president and CEO Raphael Bostic speaks on normalizing monetary policy in Madrid, Spain, for the Central Bank Series of the Global Interdependence Center.
  • Bostic explains that the Federal Open Market Committee (FOMC) began the process of normalizing its monetary policy stance three years ago with a modest increase in the policy target of 25 basis points. That gradual pace has become characteristic of the policy path since then.
  • Bostic points out that today, inflation is effectively at the FOMC's longer-run 2 percent objective and the unemployment rate has fallen to 3.7 percent, very close to the FOMC's goal of promoting maximum sustainable employment growth, or possibly some measure beyond.
  • In Bostic's view, conditions warrant the final steps in adopting a neutral stance of monetary policy. But he believes the FOMC must balance the risks of stopping short of full normalization and risking an overheated and unstable economic environment, versus going too far and short-circuiting an otherwise sustainable expansion.
  • Bostic says that to manage the balance between the risks of being too timid and being too aggressive in the Fed's monetary policy course, he can think of no superior approach than to proceed cautiously and keep a keen eye on the data.

Photo of Raphael W. BosticThank you for inviting me to speak tonight. It's an honor to be among so many distinguished colleagues from across the globe.

We are just about a week past the latest Federal Open Market Committee, or FOMC, meeting. Understandably, a lot of attention focused on what signal the Committee might send about the near-term path of the federal funds rate.

In my remarks this evening, I would like to shift the focus away from the "how many more rate moves" question. Instead, I'm going to talk about the framework that informs my thinking about U.S. monetary policy—in terms of both where we've been and the risks we might consider in defining the next phase. In particular, I'm going to address what, if any, further adjustments might be needed to get to a neutral policy.

As you may know, these comments reflect only my own opinions. Nothing I say tonight necessarily reflects the thinking of my FOMC colleagues or others in the Federal Reserve System.

Review of the path to the present
Let's begin the story in late 2015. Just about three years ago, the FOMC began the process of normalizing its monetary policy stance. The Committee moved the target range for the federal funds rate up from the near-zero floor that had been in place since December 2008. At the time, the unemployment rate stood at 5 percent. That's half the level it reached following the crisis.

The initial increase in the policy target was modest—a 25-basis-point rise—and that gradual pace has become characteristic of the policy path since then. So even with the first upward adjustment in the funds rate target, policy remained extremely accommodative.

Given the facts on the ground, continued accommodation was fully justified. In its statement and published projections, the Committee noted that there was still room for improvement in labor markets.

Perhaps more important, inflation remained a distance from the FOMC's longer-run price stability objective. At the time, core inflation was expected to come in well below 2 percent for 2015, and to not recover fully until this year. Given that the road of an economic forecaster is usually a rocky one, it is worth noting that the FOMC's outlook was amazingly prescient. On a year-over-year basis, overall inflation as measured by the personal consumption expenditures, or PCE, price index was a mere 0.3 percent by the end of 2015. The inflation rate did not consistently reach the neighborhood of 2 percent until this year.

By June 2017, the FOMC had moved its target range for the federal funds rate up by a full percentage point. The unemployment rate had fallen to 4.3 percent, far surpassing expectations. The inflation outlook had progressed roughly as anticipated. So the Committee decided to take the next step in the normalization process by announcing that it would slowly shrink the size of the Fed's balance sheet, which had more than quadrupled in the wake of the crisis.

The balance sheet normalization strategy was formulated under two guiding principles. The first was to implement the wind-down in a manner that would not disrupt financial markets. The second was to put the balance-sheet rundown in the background, if you will, focusing ongoing policy adjustments via the traditional interest rate channel. For these reasons, the Committee chose a process that was phased in, gradual, and based on a relatively fixed schedule. By most accounts, mine included, this strategy has been successful.

Which brings us to today. As I have noted, inflation is effectively at the FOMC's longer-run 2 percent objective. The unemployment rate has fallen to 3.7 percent, a level that has not been sustained for at least a half-century. So we are very close to the FOMC's other key goal—promoting maximum sustainable employment growth. Possibly, we have gone some measure beyond.

This is a record of great progress, and I believe that it is at least in part a result of the FOMC's deliberate, but patient, removal of monetary accommodation. In my view, conditions warrant the final steps in removing any remaining accommodation and adopting a neutral stance of monetary policy. The question now is as complicated as it is straightforward: What, if any, further adjustments are necessary to get to neutral?

A balance of risks
In a recent speech, Fed Chairman Jay Powell emphasized the role that risk management plays in the FOMC's decision-making. He explained how an important element of good policy is calibrating it to manage both upside and downside risk.

I couldn't agree more. As I noted in a speech I delivered last month in Baton Rouge, Louisiana: with the economy performing well and monetary policy seemingly within shouting distance of neutral, the FOMC faces both upside and downside risks with its next set of policy adjustments. As I see it, the Committee now must balance the risks of stopping short of full normalization and risking an overheated and unstable economic environment, versus going too far and short-circuiting an otherwise sustainable expansion.

I will speak to each of these risks in turn.

Too timid, too long
Is there is a risk of being too timid, for too long? Yes, and that risk would start with my earlier observation that the current level of the unemployment rate is historically quite low. Not only is the current rate at 3.7 percent but, as reflected in the Summary of Economic Projections published in September, the consensus of FOMC participants is that this level will be maintained for at least the next three years. A sustained unemployment rate of 3.7 percent or below has been seen only twice in the post-World War II era—and not at all since the late 1960s.

Perhaps more to the point, the current level of the unemployment rate is below that which the consensus of the Committee views as sustainable in the long run. The central tendency in FOMC participants' latest projections ranges from 4.3 to 4.6 percent.

Now I'll show what many would consider a disturbing picture. It highlights periods of time when the actual unemployment rate fell below what the U.S. Congressional Budget Office now estimates as the so-called natural rate of unemployment. I refer to these episodes as "high-pressure" periods.

Chart 2: Undershooting the Natural Unemployment Rate

Here is the punchline. Dating back to 1960, every high-pressure period ended in a recession. And, all but one recession was preceded by a high-pressure period.

There are many possible interpretations for the leading relationship between high-pressure periods and downturns in the economy. One is that the relationship is entirely spurious. That would be a pretty benign interpretation. I'm not convinced that such a benign interpretation is consistent with good risk management practice.

I think a risk management approach requires that we at least consider the possibility that unemployment rates that are lower than normal for an extended period are symptoms of an overheated economy.

One possible consequence of overheating is that decision-making in the private sector becomes increasingly risky, and imbalances, especially in financial markets, build up. As of yet, there does not appear to be evidence of widespread imbalances in the United States.

But I am not taking the current situation for granted. As an example, I have my eye on some recent evidence in the consumer credit market that I think bears watching. We have seen rising delinquencies on consumer credit cards at small banks. While they are not currently driving the core of the market, rising delinquencies that manifest in the small bank sector may spill out and become a broader issue. So while the aggregate picture is still a healthy one, these pockets of concern call for a more careful scrutiny of risks.

Another consequence of overheating is that inflationary pressures inevitably build up, leading the central bank, in turn, to respond aggressively, and economic weakness follows.

You might argue that the simple answer is to not respond so aggressively to building signs of inflation, but that would entail risks that few responsible central bankers would accept.

It is true that the Fed and most other advanced-economy central banks have the luxury of solid credibility for achieving and maintaining their price stability goals. But we shouldn't forget that such credibility was hard won.

Inflation expectations are stable for now, but we know little about how far the scales can tip before it is no longer so. With apologies to Joni Mitchell, when it comes to inflation expectations, "you don't know what you've got 'til it's gone." I don't think it would be a good idea to find out.

Risks of going too fast
But just as there are risks to too-timid policy adjustments, there are decidedly risks to policy adjustments that are too aggressive.

The main uncertainty, as I see it, is that it is very difficult to determine when the economy is actually overheating—especially when inflationary signals across the board remain subdued.

Consider, for example, the task of determining the "normal" rate of unemployment. My definition of normal is a sustainable level that avoids overheating in the economy. I would have pegged an unemployment rate around 5 percent as normal when the Committee started down the path of normalization in 2015. My view now is that the longer-run "normal" rate of unemployment is a good deal closer to 4 percent than 5 percent.

Estimating the capacity of the economy—about which the unemployment rate is just one measure—is extremely difficult, not least because it can be a moving target. One example is the behavior of the U.S. labor force participation rate.

Most economists accept that the secular trend of the participation rate is negative, due to the aging of the population. But the overall rate ticked up in the latest report, and has been basically flat for the past several years.

The labor force participation rate of prime-age U.S. workers, especially women, has actually been on the rise. We have no solid interpretation for this trend. But it is reasonable to suspect that larger flows of workers into the labor market, smaller flows of unemployed workers out of the labor market, and still relatively modest wage pressures are signs that there may be more slack in labor markets than typically assumed.

I found another example that capacity may be more flexible than we might think in a recent report from my staff. The Atlanta Fed collects anecdotal information from businesses throughout our district. These field reports—collected as part of what we call the Regional Economic Information Network, or REIN—have for some time included reports of acute shortages in the trucking industry.

In the most recent reports, we heard some interesting news. The recent implementation of technologies that monitor and help to enforce restrictions on how long truck drivers can be on the clock has leveled the playing field in the industry. The competitive pressures have resulted in trucking companies urging their customers to adopt processes that have resulted in less downtime as drivers wait for their cargo to be unloaded. The faster turnaround time has actually increased capacity in the industry—a result of greater efficiency rather than a greater number of trucks or truck drivers.

I don't know if this sort of development is a game changer. But it was notable to me that this anecdote came from one of the sectors that has persistently cited capacity strains. It was notable at least as a reminder that capacity is always a slippery concept.

In addition to the hazards of defining a concept like overheating, there are some rather straightforward downside macroeconomic risks. At least some of the strength in GDP growth this year can be traced to fiscal stimulus. I expect this effect to fade over the next several years. Although fiscal stimulus has clearly boosted growth in the present, I don't think we have a firm grasp on whether it is masking any building weakness for the longer run.

Add to these potential risks the uncertain outcome of global tariff and trade developments. In general, global growth seems to be slowing. Although the United States could have safely ignored these developments in the past, assuming that this is true today seems like a recipe for a policy mistake.

How, then, to manage the balance between the risks of being too timid and being too aggressive in the Fed's monetary policy course? I can think of no superior approach than to proceed cautiously and keep a keen eye on the data.

I don't think we are too far from a neutral policy, and neutral is where we want to be. We may not be there quite yet, but I am inclined to think that a tentative approach as we proceed would be appropriate.

But I am also prepared to support a more aggressive approach than my baseline view should the data deliver clearer signals of overheating. We are, for example, finally seeing some movement on wage growth. In and of itself, this is a good thing. But accelerating wage pressures would be of concern if coupled with other signs of building inflationary pressures.

I don't think I am relaying to you anything other than common sense. But what is good risk management, after all, if not the application of common sense?

Thanks for your attention. I look forward to our moderated session.