Inflation Targeting in the United Kingdom (PDF)

uncertainty regarding the state of the economy is considered by Cuckierman and Ruge-Murcia to be the driving forces of inflation bias. Non-linearities relating to the UK will be further discussed in section (3.4.4). A simple way of capturing non-linearities in policy behaviour is to estimate threshold models whereby the policy rule switches into a different regime whenever inflation breaches one or more thresholds. It may be, for example, that when inflation is in the neighbourhood of the target level, the authorities pursue a largely accommodating monetary policy, so that changes in the interest rate are more or less random since they are responding to random shocks to the economy. Once inflation rises above a given level, however, the central bank may be more aggressive in linking interest rate movements to the implicit policy rule, so that the Taylor rule best describes short-run interest rate behaviour above that level. Similarly, if inflation falls below a certain level this may generate fears of deflation and the authorities may again implement a Taylor rule - but not necessarily the same one that is employed when inflation is high. This would suggest a three-regime model.

Summary:

The last section gave an overview to the early development of inflation targeting. Taylor provides a reference path for nominal interest rates, with (observed rather than expected) output and inflation deviations from targets as feedback variables. McCallum offers a path for base money growth in consistency with hitting an (again observed) nominal GDP trajectory. They are really two sides of the same monetary coin - one defined in quantity space, the other in price space. So in principle they provide information which is complementary. The reference paths from these monetary rules serve as a consistency check on a forward-looking policy rule. Precisely because they are mechanical and based on observable variables, the rules can help to identify and quantify the discretionary input into such a rule. While non-linearities in the Taylor rule can be the result of either non-linearity in the macro-economic structure of an economy (the output-inflation trade-off ) or of asymmetry in the central bank’s preferences, it is quite likely that both of these features are present in the economy and interact to exacerbate the degree of non-linearity in the policy rule. In the eyes of Svensson, there are some disadvantages related to simple instrument rules as they were discussed in this section. He states that there is no room for the introduction of other influences on the target variables. The fact that no central bank yet explicitly introduced such a simple instrument rule is voting against its existence as direct monetary policy rule in countries that persue inflation targeting as well as in countries with other monetary policy systems. He points out that Taylor itself proposed that there might be circumstances where deviations from the rule are useful.88 So there has been research in a further direction, called target rule models.