10/14/2021 | Press release | Distributed by Public on 10/14/2021 11:08
President, Federal Reserve Bank of Richmond
Forecasters Club of New York
New York, N.Y.
Throughout the COVID-19 crisis, the Fed has supported the economy by maintaining low rates and conducting ongoing large-scale purchases of Treasurys and agency mortgage-backed securities. As the recovery has progressed, there has been a lot of talk about how and when the Fed will unwind these measures. We have attempted to give guidance on our path forward by tying our policies to specific outcomes.
Specifically, we said in September 2020 that we would reinforce our new monetary policy framework by keeping rates near zero until "labor market conditions have reached levels consistent with the Committee's assessments of maximum employment and inflation has risen to 2 percent and is on track to moderately exceed 2 percent for some time." In December, we committed to continue asset purchases "until substantial further progress has been made" toward our maximum employment and price stability goals.
Today, I want to talk about the value of issuing policy guidance tied to outcomes and share my thoughts on how to address the challenges of this approach. I hope this will be of interest to you, as forecasters. Before I dive in, I should note that the views I express are my own and not necessarily those of my colleagues on the Federal Open Market Committee (FOMC) or in the Federal Reserve System.1
The Evolution of Forward Guidance
For much of its history, the Fed was famously tight-lipped about its actions. It's hard to believe now, but until 1994, we did not publicly release a policy statement after each meeting. The minutes of FOMC meetings also remained secret until the late 1960s and then were only released with about a 90-day lag. The Fed defended this secrecy for years - Richmond's own Marvin Goodfriend analyzed many of those arguments in a 1985 paper.2 This opacity led Fed watchers to seek information wherever they could get it, including analyzing the thickness of Alan Greenspan's briefcase as he headed to FOMC meetings for clues about whether the Fed might change policy.3
We have come a long way since those days. Today, the Fed issues a policy statement immediately after each FOMC meeting, releases the minutes three weeks later, and releases verbatim transcripts after five years. This increased transparency is healthy for the Fed as a public institution, and it was also supported by a growing body of research which emphasized the importance of central bank communication and credibility.
At first, the Fed's initial steps toward greater transparency involved providing more information about current monetary policy. But in the early 2000s, the Fed also began to provide information about the likely path of future policy through "forward guidance." In 2003, the FOMC voted to reduce the federal funds rate to 1 percent and chose not to go any lower. However, the Fed still wanted to provide more accommodation to strengthen economic conditions. As argued in a key paper that same year by Gauti Eggertsson and Michael Woodford, central bank communications about the trajectory of future policy become crucial at the effective lower bound (ELB).4 Such forward guidance can allow the Fed to influence longer-term rates when it has run out of room to reduce short-term rates by setting expectations for the likely future path of short-term rates.
Initially, the Fed's forward guidance was very general. In its August 2003 policy statement, the FOMC wrote that it believed "policy accommodation can be maintained for a considerable period." The following year, when the FOMC was considering raising rates, it signaled that it would be "patient" and that the pace of rate increase was "likely to be measured." True to its word, the Fed increased rates gradually between 2004 and 2006 in increments of 25 basis points (although no one would call that pace "gradual" in today's economic environment).
During the Great Recession, the Fed again employed and continued to evolve its forward guidance. At first, the FOMC used general language similar to 2003. Then, it introduced specific, calendar-based guidance in its August 2011 statement, signaling that it would likely be necessary to maintain low rates "at least through mid-2013."
The move to calendar-based forward guidance was heavily debated at the time, as the transcripts of the FOMC meetings from that period reveal. Some policymakers felt that a calendar date helped reinforce the Fed's forecasts for the future path of the economy, which the FOMC began releasing in the form of the Summary of Economic Projections (SEP) earlier that year. But others worried that tying future policy to a date put the Fed in an awkward position. If economic conditions didn't evolve the way policymakers predicted, then they would either have to follow through on a date-based plan that no longer made sense or revise the date, diminishing its value as a signal of future behavior. In times of great uncertainty, that was asking for Fed credibility to be put on the line.
Shift to Outcome-based Guidance
In December 2012, the FOMC moved from calendar- to outcome-based guidance. It said it would be "appropriate" to keep rates low "at least as long as the unemployment rate remains above 6-1/2 percent, inflation between one and two years ahead is projected to be no more than a half percentage point above the Committee's 2 percent longer-run goal, and longer-term inflation expectations continue to be well anchored."
This made it clearer that Fed policy would be driven by economic conditions, not dates. This approach tried to provide the public with a clearer understanding of how the Fed would react to new data and it gave the Fed greater flexibility in times of heightened uncertainty.
Still, the transition to outcome-based guidance wasn't seamless. The formula outlined at that time (like the one we have outlined today) isn't a simple one. Certainly, those three criteria don't just roll off the tongue. It's also not a precise one as judgment calls are required. How would the public know if inflation expectations were no longer well anchored?
Another potential problem is that the outcomes defined could prove wrong if the economy shifts. The maximum level of unemployment, for example, moves over time, making it hard for the Fed to provide outcome-based guidance that is appropriate at all circumstances. This means that like dates, the Fed might need to revise its outcomes, leading to similar communication and credibility challenges. This is one reason why the Fed doesn't attach a specific number to the employment goal in its forward guidance today.
The Fed is facing challenges now as it seeks to navigate a highly uncertain recovery from the pandemic. Inflation has been above our long-run 2 percent target for months, but will this run-up in prices be transitory or sustained? Unemployment remains above pre-pandemic levels, but has the economy changed in ways that have shifted the maximum level of employment? What does "substantial further progress" look like for a broad-based and inclusive metric that has many measures? Should fiscal policies not imagined when our guidance was defined change its terms?
Finally, it seems clear to me that many audiences find outcome-based guidance unsatisfying because it cannot provide a definitive roadmap of the Fed's future policy path. Trading instruments are often date-based, so traders would prefer to know exactly when monetary policy is going to change. For reporters and the public they serve, outcome-based guidance can seem like inside baseball and is not as easy to process as dates. You as forecasters get this, I'm sure.
I think this has led reporters and market analysts to overemphasize the SEP, particularly the "dot plot" that presents FOMC members' individual estimates of the appropriate future path of policy. The median forecast is often treated as a commitment by the Fed to a particular date-based path. But as Chair Powell has made clear in his press conferences, the SEP is a collection of individual forecasts, not a committee consensus. It can't be a substitute for the forward guidance the FOMC presents in its policy statement.
Where Do We Go From Here?
If the public keeps asking for dates, should the Fed go back to issuing calendar-based forward guidance? It's clear from past experience that this isn't the optimal path. Markets and reporters may want clear dates, but in times of high uncertainty, the Fed can't credibly commit to guiding policy by dates rather than data.
Sticking with outcome-based guidance, the Fed could try to be more specific about its thresholds, but past experience also suggests this approach has limitations. It's harder to get alignment among the committee. The higher the level of specificity, the higher the risk that you'll bind yourself to a suboptimal path. Surely, there's some value to leveraging good judgment. In addition, some of the Fed's objectives, like maximum employment, are hard to forecast and are influenced by factors outside of our control.
We could strengthen the connection between the SEP and outcome-based guidance. I like the SEP because it disciplines me to tie my policy prescription to my economic forecast. In times when forward guidance is a crucial component of Fed communications, I think that through very carefully. Currently, however, the SEP does not tie economic forecasts to a dot on the dot plot. Doing so would provide a clearer picture of each FOMC member's individual reaction function, and taken as a whole, this could help shed more light on the Fed's overall reaction function. That said, since the SEP isn't a committee consensus, we could still run into a problem where the SEP and policy statement send conflicting messages.
Ultimately, I think the most important thing we can do to build confidence in forward guidance is to cleanly execute. In the early 2000s, the Fed signaled that it would follow a gradual path for rate liftoff and then did so. The 2013 taper tantrum was an example of when Fed communications and forward guidance were not so well aligned.
Hopefully, we are executing during the COVID-19 recovery in a way that builds credibility. Regarding our guidance on asset purchases:
That still leaves rate forward guidance, which is explicitly different. We have hit 2 percent on inflation, but we still have a lot to learn about whether recent inflation levels will be sustained and how much room we have to run in the labor market until we get to maximum employment. As COVID-19 hopefully eases, I expect the answers to these questions to become clearer.
Cleanly executing communication going forward is my goal. Doing so best cements outcome-based guidance as a tool comfortable for us and valuable for the market, for the public, and for you as forecasters.
Thank you to Tim Sablik for assistance preparing these remarks.
Marvin Goodfriend, "Monetary Mystique: Secrecy and Central Banking," Federal Reserve Bank of Richmond Working Paper No. 85-7, Revised August 1985.
William T. Gavin and Rachel J. Mandal, "Inside the Briefcase: The Art of Predicting the Federal Reserve," Federal Reserve Bank of St. Louis Regional Economist, July 2000.
Gauti B. Eggertsson and Michael Woodford, "The Zero Bound on Interest Rates and Optimal Monetary Policy," Brookings Papers on Economic Activity, 2003, no. 1.