Thursday, February 26, 2015

The Taylor rule conundrum

Back in February 2005 Alan Greenspan referred to the abnormal (low) level of US long-term interest rates as a conundrum:

"For the moment, the broadly unanticipated behavior of world bond markets remains a conundrum. Bond price movements may be a short-term aberration, but it will be some time before we are able to better judge the forces underlying recent experience."

A month later, Ben Bernanke, proposed the idea of a global saving glut as the main reason for low long-term real interest rates. In a world where capital markets are global, interest are determined by global forces and not by domestic macroeconomic conditions.

This behavior is also very much related to the discussion around "global liquidity" and the potential influence of monetary policy in the US on monetary policy conditions in emerging markets. The difference is that in this case we are talking about short-term rates where we typically expect more control by the central bank and a stronger correlation domestic conditions.

To illustrate this point let me use a slide from a presentation two days ago by Hyun Song Shin (BIS) at the Bank of England discussing the future agenda of central banks. The slide includes the following chart (click on it for a larger image).

The chart shows the behavior of emerging markets central bank interest rates in comparison with a standard Taylor rule. The chart shows that since 2000, central banks have set interest rates significantly below the level implied by the Taylor rule (the same behavior is also true among advanced economies although not to the same extent). If one looks carefully at the scale, interest rates are 6 to 8 percentage points lower than those implied by a Taylor rule. This would imply an incredibly expansionary monetary policy. The chart comes from a paper by researchers at BIS (Hofmann and Bogdanov) who on their discussion of this result they argue that

"This finding suggests that monetary policy has probably been systematically accommodative for most of the past decade. The deviation may, however, in part also reflect lower levels of equilibrium real interest rates that might introduce an upward bias in the traditional Taylor rule."

Monetary policy can be too accommodative when central banks follow US interest rates to avoid appreciations of their currencies. But if monetary policy was that accommodative for more than a decade we should have seen increasing inflation rates during those years. That was not what we saw, inflation rates remain stable (and even decreasing) in most of these markets. Today, where interest rates remain very low compared to those implied by the Taylor rule, we talk about global deflation, not global inflation.

So it must be that the fundamental cause must be related to lower levels of equilibrium real interest rates and these are determined by global forces (otherwise why would all countries behave in the same way). Interestingly, the deviations from the Taylor rule coincide with the period where global imbalances started.

So here is yet another interest rate conundrum, this time related to short-term interest rates. How much do central banks control short-term rates in a world where capital markets are global? How relevant is a Taylor-rule approach to analyze the appropriateness of central bank interest rates?

Antonio Fatás