944 Monetary Policy, Beliefs, Unemployment and Inflation
Real exchange rate (RXR). Effective nominal rates, using the 2000 patterns of
trade matrix, deflated by the consumer expenditure deflator.Source: National Institute database.
Acknowledgments
Thanks are due to Ali Al-Eyd, Joe Pearlman and Malcolm Pemberton for their considerable
help with the analysis in section 18.5. I would also like to thank Ray Barrell, Alan Budd,
Richard Dennis, Mike Dicks, Stephen Hall, Richard Jackman, Joe Pearlman, Martin Weale and
the editors of the Handbook, Terence Mills and Kerry Patterson, for their helpful comments
on an earlier draft of the chapter. Simon Kirby provided crucial help at several stages in the
preparation of the chapter. All the aforementioned are absolved from responsibility for the
contents of this work, which is the sole responsibility of the author.
Notes
- The inflationary increases of 1979–80 were also exacerbated by the switch from direct to
indirect tax in the Budget. Further detail are given in section 18.5.2.4. - A recent clear review of this methodology is found in Perron (2006).
- But see Kara and Nelson (2003) for an example.
- In what follows, expectations are taken to be conditional on information available at
periodt. - Surico (2006) echoes this view when estimating the NKPC which, he argues, needs to
allow for the interest rate reaction function actually in force at the time. - This uses the same notation as SW (2002) and follows their convention of ignoring
the intercepts in the equations for notational convenience, though these are used in
estimation. - The standard interpretation is that forward-looking expected variables enter in their
lagged representation. - The so-called “timing protocol” is largely standard though, whereby the government
first sets the systematic part of inflation and then the public form their expectation for
inflation. - Perhaps the most sophisticated example of this class of studies is that of Svensson (2000).
- This is largely standard in the cited examples. The present version draws on the Tetlow
and von zur Muehlen (2001) account. - Note that throughout this chapter, the terms “natural rate” and “NAIRU” are used inter-
changeably. The switch from one to the other, though possibly confusing to the reader,
is made to accord with what is used in the original papers. - This version of the AS equation, with unemployment as the dependent variable, is stan-
dard in the Beliefs literature. We will refer to it throughout the rest of the chapter as the
Phillips curve, hopefully without risk of ambiguity. - There are some differences in the precise methods used in the literature. Primiceri (2005),
for example, uses Kalman filtering. - See Ellison and Yates (2007) for the links between the model’s mean dynamics, their
stability conditions and the SCE. - To establish them a further optimal control problem is used where mean dynamics are
perturbed and a weighting function is used that measures the likelihood of the shocks
needed to perturb beliefs. For details of this, see Cho, Williams and Sargent (2002). - Section 18.5 discusses how the “inflation surprise” term is included in the model.