How useful are interest rates?

Alicat/E+ via Getty Images

So if the Fed raises interest rates, how much and how soon will that contribute to inflation? For another project, I went back to classic Valérie Ramey criticism. Here is his replica and update of two classic estimates:

interest rate

Two estimates of the effect of monetary policy shocks. Top: Christiano et al. (1999) identify. Full specification 1965m1-1995m6: solid black lines; Full specification 1983m1-2007m12: short dashed blue lines; 1983m1-2007m12, omits silver and reserves: long dashed red lines. The light gray bands are 90% confidence bands. Bottom: Romer and Romer monetary shock. Coibion ​​VAR 1969m3-1996m12: solid black lines; 1983m1-2007m12: short dotted blue lines; 1969m3-2007m12: long dashed red lines (Ramey (2016))

The left side tells us what the fed funds rate typically does after the Fed raises it. The right shows the effect of the rate hike on level of the CPI. Inflation is the slope of the curve. The horizontal axis represents quarters. The top panel uses vector autoregression. The bottom panel uses the Romer and Romer reading of the Fed minutes to isolate a monetary policy shock.

Upper pane (VAR): Multiplying by 10, a 2 percentage point rise in the fund rate (blue dash) could reduce cumulative inflation by one percentage point in three years (12 quarters) before it hits. runs out of steam. The black line is the most promising, but this is essentially the experience of 1980. Yet, multiplying by 5, a 2 percentage point increase in the fund rate only lowers inflation by half a percent during these first three years (12 quarters), although after 10 years (40 quarters) you get a full reduction in the price level.

Bottom panel (Narrative): In the black and red lines that include the 1980 shock, a 3% rise in interest rates produces no noticeable fall in inflation for the first three years. 10 years later, the price level is 4% lower, but this represents a reduction of 0.4% per year in inflation. The blue lines that exclude 1980 show a plausible shock of longer duration, but a 1% higher interest rate produces only a 1% lower price level in 10 years, or 0.1% per year.

The problem is the short-lived Phillips curve, which I emphasized in my WSJ op-ed. In VARs, the Fed is pretty good at inducing a recession. Here are the effects of Romer-Romer shocks on output and unemployment:

production and unemployment


It’s just that inducing recessions isn’t particularly effective in reducing inflation.

And I chose beautiful graphics. Many estimates do not find any effect on inflation or even positive:



No theory today, just facts. This is the empirical basis for the idea that the Fed can quickly stop inflation by raising interest rates. The underlying machinery has done its best for 50 years on the subject to separate causality from correlation and to insulate Fed actions from other influences on inflation. Perfect, no, but that’s what we have.

Maybe relying on the Fed to stop inflation on its own isn’t such a good idea. And I don’t have any more Jaws and WIN buttons in mind.

Original post

Editor’s note: The summary bullet points for this article were chosen by the Seeking Alpha editors.

About Catriona

Check Also

The mythical soft landing seems slightly more achievable

Comment this story Comment Financial markets’ sudden reaction to Friday’s payroll data was a classic …