Tuesday, 23 May 2017

Blog

The Taylor Rule and the natural rate of interest

April 20, 2015

A few weeks ago I commented on whether or not the Market Monetarist position that a 5% growth rate of NGDP (before the subprime crisis) could have been too much. The main point was that other nominal variables did not behave in accordance to what would expected in monetary equilibrium. Because other widely used measures, like Taylor’s rule, point in the direction of monetary policy being “too loose,” I think this is a valid question to raise to the NGDP Targeting rule so that not only the policy conception is sound, but that the target is also correctly chosen.

Given that monetary policy is done through the credit markets, the interest rate is a key variable. An important interest rate is the so called “natural rate of interest.” A specific definition of the natural rate might depend on the model being used and assumptions held, but Wicksell’s natural rate points to long-run stability. The natural rate of equilibrium is such that does not produce a change in the relative price of future goods (capital goods) to present goods (consumer goods). The Federal Reserve, then, does not want to permanently deviate from the natural rate of interest.

The natural rate of interest is, however, unobservable. But there are estimations. Two of them are the calculations by Laubach and Williams (2003) and by Selgin, Beckworth, and Bahadir (2015). Two things, then, can be done. One is to compare the Federal Funds rate (which is affected by the Fed’s monetary policy) with the natural rate of interest. The second one is to calculate a modified Taylor Rule. This modified Taylor Rule prescribes adjustments to the short-run interest rates when there are deviations from potential output and the target inflation not to an assumed fix 2.5% real interest rate plus inflation, but to the estimated natural rates of interests.

The figures blow show the Federal Funds rate along to the natural rate estimations and the Classic Taylor rule next to the modified Taylor Rules using each one of these natural rate estimations.

 

graph1

 

graph2

If we take these estimations as valid, then the graphs suggests that the Federal Fund rate was too low between 2002 and 2004/05. But the second graph also shows that the classic Taylor Rule prescribes a higher Federal Funds rate level than it should.

Here is the updated version of the paper (without estimations by Selgin, et.al.)