This thesis presents a structural framework which accounts for two key empirical phenomena in monetary economics: the ‘Taylor principle’ and the ‘Fisher relation’. The former suggests that there exists a greater-than-proportional relationship between the nominal interest rate and inflation in the short-run and the latter implies that a one-for-one relationship holds at lower frequencies. Although these relationships do feature in the ubiquitous, ‘cashless’ New Keynesian framework, it has been suggested that monetary variables are required in order to render this model ‘complete’ (e.g. Nelson, 2008a). Chapter-I demonstrates that an ‘implicit’ interest rate rule can be derived as a general equilibrium condition of models in which the central bank adheres to a money growth rule. Chapter-II compares the equilibrium condition of a standard cash-in-advance model to the interest rate rule of Taylor (1993) for a post-war sample of U.S. data. However, we demonstrate that in order to replicate the Taylor principle, the underlying model must be generalised to allow the velocity of money to vary. We use the model of Benk et al. (2008, 2010) to do so and show analytically that the resulting ‘implicit rule’ features the requisite greater-than-proportional relationship. Chapter-III applies standard econometric techniques to simulated data obtained from the Benk et al. model and the estimates obtained offer support for this theoretical prediction. Chapter-IV establishes that the Fisher relation emerges when low frequency trends in the simulated data are retained and under a related ‘long-run’ implicit rule. Chapter-IV also considers the post-war sample of U.S. data analysed in Chapter-II. While disparate empirical literatures have obtained evidence for both the Taylor principle and the Fisher relation, we show that these results can be obtained from a unified theoretical framework. Several restricted empirical specifications further suggest that standard interest rate rules which omit monetary variables might provide biased coefficient estimates.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:590362 |
| Date | January 2013 |
| Creators | Davies, Ceri Rees |
| Publisher | Cardiff University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://orca.cf.ac.uk/57432/ |
Page generated in 0.0019 seconds