Some of the data in the figure comes from DR’s table 1 showing the number of basis points (hundredths of a percent, so 100 bps is one percentage point) that the Fed has reduced the main tool it controls–the Federal funds rate–over a number of recessions. The black dot shows where the funds rate was when they began their lowering campaign, the arrows show how much they lowered, and the green box shows where they stopped (this excellent graphic was custom made for me by Ben Spielberg; see data note below). For example, to apply countercyclical monetary policy in the case of the Great Recession–the last bar in the figure–the Fed took the rate down from about 500 bps to zero.
The last dot shows where the rate is today–close to zero (~40 bps)–which is where it should be IMHO as we’re not yet at full employment and there’s no worrisome signs of overheating; inflation remains quiescent such that the Fed keeps missing their 2% inflation target on the downside.
DR’s simulations assume that last dot climbs in time to give the Fed some height to drop from when the next downturn hits (importantly, he stresses that the neutral funds rate is very likely lower than it used to be), but, as I argue in the piece, with some evidence from market expectations of the funds rate, I’m skeptical.
Data note for graph: The maximum and minimum federal funds rates before 1990 were almost all calculated using the methodology described in Reifschneider’s footnote 1, table 1. They are the maximum and minimum effective federal funds rates in any given month spanning from 6 months before the recession began to 6 months after the recession ended, with only one exception: the end period extends to only the official end of the 1980 recession in July of 1980, and not 6 months afterwards, because rates began rising afterwards and including those months would have made the drop appear larger than it actually was. During the three most recent recessions, the time periods used to determine the maximum and minimum effective federal funds rates were June 1990 to December 1992 (DR’s ftnt has January 2002 for the latter date for this period but we assume that’s a typo), December 2000 to January 2002, and August 2007 to December 2008. The methodology for the post-1980 recessions is slightly different than that Reifschneider used, as he uses “intended” fund rates beginning in 1990, which differ slightly from effective rates, but they tell the same story.