Yesterday the Fed published a paper that provided some indication of what the fan charts for the March FOMC meeting will look like. The minutes of the February 2017 FOMC meeting mentioned a forthcoming “staff paper on measures of forecast uncertainty in the SEP, including those that would be used as the basis for fan charts in the SEP.”
The paper makes several key points:
∙ Considerable uncertainty surrounds FOMC participants’ macroeconomic projections. ∙ Indeed, forecast errors are an inevitable part of the forecast process. Thus the adage: “It’s tough to make predictions, especially about the future” (Attributed to various people, including Niels Bohr and Yogi Berra.)
∙ Past prediction errors are taken to be a rough guide of forecast errors that might occur in the future. ∙ The paper looks at forecast errors by a number of forecasters: FOMC participants, the Board staff, the CBO, the CEA, the Blue Chip survey, and the Survey of Professional Forecasters. Forecasts from each of these sources were found to have similar degrees of accuracy.
∙ Estimates of uncertainty surrounding GDP/employment forecasts and rate paths were large, and are now considerably larger following the financial crisis. In contrast, less uncertainty surrounds inflation forecasts, and there has not been much change following the financial crisis.
∙ Many central banks have found that fan charts are an effective way of communicating the uncertainty around their own forecasts. The fan charts developed here by Board staff show uncertainty bands constructed so that outcomes will fall within that range 70% of the time, subject of course to certain assumptions.
∙ Fan charts may help the public better appreciate that the FOMC’s actual policy generally will not correspond to what it previously projected—that is, the median dots.
∙ Symmetric uncertainty bands around the median fund’s rate projections constructed in this mechanical way are problematic because they might be interpreted as indicating that the FOMC is prepared to lower rates into negative territory, even if participants are not inclined to do so.
Below we show fan charts for the September 2016 SEP from the Fed’s recent paper on forecast uncertainty. Note that the February 2017 FOMC minutes stated that the fan charts “would illustrate the uncertainty that attends participants’ macroeconomic projections,” which suggests that uncertainty estimates for the median projected funds rate path will not be published.
Yesterday’s paper notes:
The lower and upper bounds of the typical range of possible outcomes equal the median projections from the September 2016 Summary of Economic Projections (SEP) plus (upper) or minus (lower) the average historical root mean squared prediction errors for each series at the appropriate forecast horizon. If future prediction errors are stable, unbiased, and normally distributed, approximately 70 percent of future outcomes
can be expected on average to fall within the upper and lower bounds, with remaining outcomes falling symmetrically above and below the bounds.