Global Central Bank Focus

Data Dependence Is Not a Monetary Policy But Are the Dots

The future path of the policy rate will depend on both the future data and the Fed’s reaction function to that data.

Please find below for your consideration a selection of quotations from Federal Reserve Chair Janet Yellen’s press conference following the most recent meeting of the Federal Open Market Committee. Note how these passages share a common refrain:

“ … although policy will be data dependent, economic conditions are currently anticipated to evolve in a manner that will warrant only gradual increases in the target federal funds rate.”

“ … the appropriate policy decision is going to be data dependent, and all of us will be looking at incoming data. And our opinions about the appropriate timing of normalization are likely to shift as we look at how the data evolves.”

“ … policies should be data dependent, and the Committee is always doing its best to assess the implications of incoming data.”

“ … when the people write down their dots in the SEP [Summary of Economic Projections], they’re making forecasts about what unfolding data is likely to show. But the participants will all be – their views will evolve with unfolding data.”

“ … as I’ve said previously and as we’ve said in the statement, it will depend on a wide range of data and not on any simple indicators.”

“ … we will be making decisions, however, that depend on the actual data that we see in the months ahead. So, certainly, we could see data in the months ahead that will justify the expectations that you see in the so-called dot plot. But, again, the important point is, no decision has been made by the Committee about what the right timing is of an increase. It will depend on unfolding data in the months ahead.”

“We will be responding to incoming data. We’ve tried to make that clear.”

As expected, the Fed at its June meeting decided not to hike interest rates and begin the process of policy normalization, but the committee, via the chair, did want us to know that when they do, they will be … data dependent!

What does this mean, and why do they want us to know it? Is data dependence in and of itself a monetary policy strategy? If not, then what kind of Fed strategy would be consistent with data dependence? And finally, what about those dots?

The evolution of Fed communication: from calendar guidance to data dependence

Since 2003, the Fed has often communicated its intentions for future monetary policy with reference to the passage of time – a strategy known as “calendar guidance.” (For a detailed discussion, see our November 2013 Global Perspectives, “Guidance Counselors.”) For example, in 2003 the Fed said that rates would be on hold “for a considerable period,” and then in 2004–2006 that they would be normalized at a “measured pace.” Fast forward to 2011: Facing a weak recovery from the worst recession since the 1930s and constrained by a fed funds rate pinned just north of the zero bound, the Fed began to offer very specific calendar guidance regarding the passage of time before it anticipated it would even consider raising rates. (See the October 2011 article, “Eyes on the Prize.”) For example, in September 2011 the Fed statement said this:

“The Committee also decided to keep the target range for the federal funds rate at 0 to 1/4 percent and currently anticipates that economic conditions – including low rates of resource utilization and a subdued outlook for inflation over the medium run – are likely to warrant exceptionally low levels for the federal funds rate at least through mid-2013.”

But while financial markets like calendar guidance, the Fed doesn’t, and so in December 2012 it began to replace calendar guidance with something else: guidance based on observable macroeconomic outcomes without specific reference to calendar dates. For while there may be a strong case for calendar guidance when the policy rate is pinned at the zero bound, the perceived costs (to the Fed) start to rise as policymakers anticipate that the Fed will at some point begin to consider the process of normalizing interest rates. (See the February 2014 article, “Is It Time for the Fed to ‘Level’ With Markets?”) Here is an example of specific outcome guidance from the January 2014 Fed statement:

“The Committee [expects] that the current exceptionally low target range for the federal funds rate of 0 to 1/4 percent will be appropriate 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.”

Over time, the original formulation of outcome guidance lost its relevance as the unemployment rate plunged faster and farther than the Fed initially expected. But instead of modifying the thresholds, the Fed in March 2014 simply dropped any mention of guidance based on macroeconomic thresholds from its statement.

In recent statements, the Fed has replaced threshold guidance with language stating that, before it raises rates, it wants to see “continued progress” toward improvement in labor market conditions while being “reasonably confident” that inflation in coming years will begin to rise toward the 2% target. And although the statement itself has been silent about the pace of rate hikes once normalization begins, Chair Yellen in her press conference echoed what other officials have been saying – that the pace of rate normalization is likely to be gradual:

“ … although policy will be data dependent, economic conditions are currently anticipated to evolve in a manner that will warrant only gradual increases in the target federal funds rate.”

Why data dependence is not in and of itself a monetary policy
Although a data-dependent Fed is one that appears to retain a great deal of optionality on the timing and pace of future rate moves, data dependence itself is not a monetary policy. To see why, consider that the level of aggregate demand in the economy at any time depends on the bond yields (which influence investment spending and housing construction), stock prices (which influence household spending through the wealth effect) and exchange rate (which influences export demand) that prevail at that time. And all three of these factors depend in part on market expectations for the future path of the overnight policy rate. But the future path of the policy rate will depend on both the future data and the Fed’s reaction function to that data. Can we infer anything about the Fed’s reaction function from the “dot plot?”

Dots are not enough
The dots provide a lot of information, but not enough to reconstruct the FOMC reaction function. Each dot represents the preferred policy rate of each participant at each date given her or his individual forecast of the economy. The dot is not the mean of her or his distribution but the most likely, or modal, outcome. Not all dots are created equal in that only some participants actually vote at each meeting. Also, Fed decisions are not by unanimity, so dissenting views, if in a minority, do not affect the outcome. Finally, the chair’s influence on the policy decision likely well exceeds her single vote out of the 12 FOMC members who vote when the committee is at full strength. So only with additional assumptions can one see past these imperfections to infer an FOMC reaction function from the dots. Typically this is done by focusing on the median dot and aligning it with the median FOMC view for unemployment and inflation embodied in the SEP. So if the economy plays out as the Fed baseline projects, then under these assumptions, the path for normalization could play out as indicated by the median blue dot.

From dots to optimal control
But crucially, the dots alone don’t tell us how policy will play out if the macro data evolve differently from the baseline. For that, we still need to ascertain the FOMC reaction function. Figure 1 shows one attempt to do that by comparing the path for the median blue dot to the path for policy if the Fed follows an optimal control policy reaction function and the data evolve according to the median of the SEP. The two paths track each other reasonably closely assuming a lift-off later this year and the pace of hikes implied by the median blue dot. By 2017, they converge, with the blue dot path moving modestly higher than the optimal control path.

So will the Fed react to macro data as the optimal control path suggests? The Fed isn’t saying, but there are reasons to believe it takes the optimal control reaction function seriously. When Yellen was vice chair, she gave two notable speeches that highlighted the optimal control reaction function as one way to think about policy normalization. In fact, this reaction function may well offer the best insight we have about how Fed policy will depend on the evolution of data. So while data dependence cannot in and of itself account for and predict future Fed monetary policy actions, the dots and the SEP projections, under certain assumptions, suggest that optimal control may be able to.

The Author

Richard Clarida

Global Strategic Advisor

View Profile

Latest Insights


All investments contain risk and may lose value.  This material contains the opinions of the author but not necessarily those of PIMCO and such opinions are subject to change without notice. This material has been distributed for informational purposes only. Forecasts, estimates and certain information contained herein are based upon proprietary research and should not be considered as investment advice or a recommendation of any particular security, strategy or investment product. Information contained herein has been obtained from sources believed to be reliable, but not guaranteed. No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission. PIMCO and YOUR GLOBAL INVESTMENT AUTHORITY are trademarks or registered trademarks of Allianz Asset Management of America L.P. and Pacific Investment Management Company LLC, respectively, in the United States and throughout the world. ©2015, PIMCO.