Monetary Policy and the FOMC’s Economic Projections

The Federal Reserve has further increased its transparency over the last couple of years. In 2007, for example, the Federal Open Market Committee (FOMC) introduced the Survey of Economic Projections (SEP), which reports Committee participants’ projections for GDP growth, unemployment, and inflation. In January of 2012, the projections were expanded to include the federal funds rate. These projections are based on each participant’s view of appropriate monetary policy.

The inclusion of interest rate projections allows a rare opportunity to see whether a simple “guide post” might accurately describe participants’ views on appropriate policy. Monetary policy is frequently discussed in terms of guideposts, and often these are presented in the form of Taylor-type rules.

The original Taylor rule posited that the current federal funds rate is set as a function of the long-run interest rate, deviations of inflation from the FOMC’s target (currently 2 percent), and deviations of economic output from its potential. One common modification of this rule, which is more consistent with the Committee’s dual mandate of promoting price stability and maximum employment, is to look at deviations of unemployment from long-run unemployment instead of GDP from its potential. Until interest rates hit near-zero and could not be lowered any further, this rule tracked the actual funds rate fairly closely.

We look at how this rule lines up with the Committee’s statement that the current extraordinary monetary policy accommodation will continue until late 2014. Since the Committee’s statement reflects the consensus opinion, we will also see how well the rule does in describing the entire distribution of Committee members’ interest rate projections, reported in the SEP.

First let us summarize the FOMC participants’ stated views. In both January and April, the Committee’s statement said “economic conditions are likely to warrant exceptionally low levels of the federal funds rate at least through late 2014.” Similarly, in the SEP released after those meetings, the median Committee member set the time for leaving the zero lower bound—“liftoff”—somewhere during that year. While two participants expected this hike to occur in 2016 at the January meeting, by April there were no participants expecting the hike to occur that late.

Looking at individual projections for real economic variables, we see that in April, FOMC participants were expecting lower unemployment and higher inflation in the short term than they were in January. Yet by 2014, the forecasts are largely unchanged.

Economic Projections of FOMC Members, January and April 2012

Variable		Central tendency				Range
		2012	2013	2014	Longer run	2012	2013	2014	Longer run
Percent change in real GDP	April	2.4–2.9	2.7–3.1	3.1–3.6	2.3–2.6	2.1–3.0	2.4–3.8	2.9–4.3	2.2–3.0
	January	2.2–2.7	2.8–3.2	3.3–4.0	2.3–2.6	2.1–3.0	2.4–3.8	2.8–4.3	2.2–3.0
Unemployment rate	April	7.8–8.0	7.3–7.7	6.7–7.4	5.2–6.0	7.8–8.2	7.0–8.1	6.3–7.7	4.9–6.0
	January	8.2–8.5	7.4–8.1	6.7–7.6	5.2–6.0	7.8–8.6	7.0–8.2	6.3–7.7	5.0–6.0
PCE inflation	April	1.9–2.0	1.6–2.0	1.7–2.0	2.0	1.8–2.3	1.5–2.1	1.5–2.2	2.0
	January	1.4–1.8	1.4–2.0	1.6–2.0	2.0	1.3–2.5	1.4–2.3	1.5–2.1	2.0
Core PCE inflation	April	1.8–2.0	1.7–2.0	1.8–2.0		1.7–2.0	1.6–2.1	1.7–2.2
	January	1.5–1.8	1.5–2.0	1.6–2.0		1.3–2.0	1.4–2.0	1.4–2.0

Source: Bureau of Economic Analysis.

Next, we use the economic projections from the January and April SEP to produce a federal funds rate path into the future using the Taylor rule discussed earlier. We then compare the federal funds rate path implied by the Taylor rule to the liftoff dates implied by participants’ interest rate projections.

We assume that the Committee participant predicting the earliest hike in the interest rate has the highest long-term interest rate projections, the lowest unemployment projections, the highest long-run unemployment rate projections, and the highest inflation projections. This is necessary since we do not have the data to map which inflation rate goes with which unemployment rate, interest rate, etc. We do this for the upper and lower ranges, the upper and lowest central tendency of the projections (which excludes the three highest and three lowest projections for each variable in each year), and the median path or the midpoint of the projections.

According to the Taylor rule paths, the first fed funds rate increase for the median Committee participant would be in the second quarter of 2014 for January and the fourth quarter of 2013 for April. Recall that the median Committee participant set a liftoff date of 2014 in both the January and April SEP.

Exit Timing: Taylor Rule versus Projections: 50-Basis Point Cutoff for Liftoff

	Timing of the first rate increase
	Bottom of range	Bottom of central tendency	Median	Top of central tendency	Top of range
January 2012 SEP	2012	2013	2014	2015	2016
January Taylor rule	2012:Q3	2013:Q1	2014:Q2	2016:Q2	2017:Q1
April 2012 SEP	2012	2013	2014	2015	2015
April Taylor rule	2012:Q3	2012:Q3	2013:Q4	2015:Q3	2017:Q1

Note: The central tendency excludes the three highest and three lowest projections for each variable in each year. The range includes all participants’ projections, from lowest to highest, in that year.

Sources: Federal Reserve Bank of Philadelphia; Summary of Economic Projections, January 2012 and April 2012, Federal Reserve Board; Bureau of Economic Analysis; Bureau of Labor Statistics; authors’ calculations.

We can expand this exercise to the rest of the projection distribution as well. If the rule perfectly describes Committee participants’ views of appropriate monetary policy, the distribution of projections produced by the rule would match the distribution of projections submitted by Committee participants. That is, the rule’s projections would lie on the 45-degree line when plotted against the participants’ submitted predictions. The Taylor rule, at least for the median Committee participant, does a good job in matching the first projected liftoff date.

A simple way of measuring how well the rule does in matching individual participants’ interest rate projections is to look at how far apart the projected liftoff dates are at each point in the distribution. According to the Taylor rule paths, the first fed funds rate increase for the median Committee participant would be in the second quarter of 2014 for the January SEP and the fourth quarter of 2013 for April. Recall that the median Committee participant set a liftoff date of 2014 in both the January and April SEP. We measure the differences, or the misses, in terms of quarters. In both January and April, the average miss at the five points of the distribution was 2.6 quarters. Interestingly, the rule tended to predict later liftoff dates than the dates that were submitted in the projections. In January, the rule predicted a liftoff date that was an average of 1.4 quarters later than the dates submitted by Committee participants. By April, it had decreased to 1.0 quarter, though the difference remained.

Exit Timing Differences: Taylor Rule versus Projections

	Quarters apart
	Average miss	Average absolute miss
January 2012	1.4	2.6
April 2012	1.0	2.6

Sources: Federal Reserve Bank of Philadelphia; January and April 2012 SEP; Bureau of Economic Analysis; Bureau of Labor Statistics; authors’ calculations.

For the upper ranges, the liftoff-date miss is quite significant. In January, the latest liftoff date in the submitted fed funds rate projections occurred 4 quarters earlier than the latest liftoff date implied by the rule. The difference was an incredible 7 quarters in April. This is actually to be expected since we assumed that the Committee participants projecting the latest funds rate hike had the highest unemployment projections, the lowest long-run unemployment rate projections, and the lowest inflation projections. We would expect that the projections based on the rule for the upper range would be later because we are putting all of the extremes into one projection. The rule-based ranges are therefore an upper limit on the timing of the exit date implied by the Taylor rule. For the lower ranges, we would anticipate the opposite. That is, we would expect the slope of the dots in the scatter plot to be flatter than the 45-degree line. This is exactly what we see.

Even then, we should not expect a perfect fit. First, Committee participants report projections only at an annual basis. To be more consistent with a quarterly Taylor rule, the interest rate projections for this analysis are linearly extrapolated from one year to the next. Second, we make the strong assumption that liftoff occurs when the projected interest rate paths exceed 50 basis points. Third, even with these caveats, it should be obvious that no Committee participant would truly think that appropriate monetary policy would be to slavishly follow such a rule. There are myriad other factors that Committee participants would also consider.

Nevertheless, this exercise illustrates that such a rule roughly captures many Committee participants’ views of appropriate monetary policy. It suggests that if the economic data continues to improve, the projected liftoff dates will be pushed sooner, and the language of the Committee will likely follow suit.