TESTING SUPEREXOGENEITY: THE DEMAND FOR BROAD MONEY IN THE UK

AuthorA. S. Hurn,V. A. Muscatelli
Published date01 November 1992
DOIhttp://doi.org/10.1111/j.1468-0084.1992.mp54004004.x
Date01 November 1992
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 54,4(1992)
0305-9049 $3.00
TESTING SUPEREXOGENEITY: THE DEMAND
FOR BROAD MONEY IN THE UK
A. £ Hum and V A. Muscatelli
I. INTRODUCTION
The relevance of the 'Lucas critique" to the problem of modelling the
demand for money had received a great deal of attention in the UK empirical
literature. To date, the debate has focused principally on the issue of whether
the Lucas critique is applicable in the case of the demand for Ml in the UK.
There have been numerous successful attempts to fit forward-looking buffer-
stock models of the demand for Ml in the UK (cf Cuthbertson and Taylor,
1987, Cuthbertson, 1988, Muscateffi, 1988). The problem in assessing the
validity of such forward-looking interpretations of the demand for money is
the well-known obstacle of observational equivalence: the reduced forms of a
forward-looking model and a more conventional backward-looking error-
correction model coincide. Hendry (1988) suggests a test to discriminate
between the two hypotheses. He argues that the stability of the backward-
looking (conditional) model for Ml in the UK in comparison to the instability
of the marginal models for income, the interest rate, and the price level,
indicates a refutation of the forward-looking model.2 Cuthbertson (1991)
argues that the interpretation offered by the forward-looking model for Ml
may still be valid in cases where the econometrician has failed to identify the
'true' constant-parameter expectations-generating equation. Of course, this
merely states that the power of any testing procedure to discriminate between
the two interpretations depends critically on the existence of policy regime
shifts.The evidence on Ml in the UK does not lend support to the Lucas
critique.3 One possibility (suggested, inter a/ia by Bain and McGregor, 1985,
1See Lucas(1976).
2alternative approach to the problem of discriminating between forward-looking and
feedback-only models of the demand for Ml is followed by Muscatelli (1989), who employs
variance-encompassing tests in order to discriminate between these two types of models, and
seems to find evidence in favour of the error-correction model (for a critique of these tests, see
Cuthbertson and Taylor, 1991). However, as noted by Favero and Hendry (1989) and Favero
(1991), there are problems in obtaining an unambiguous error-variance ranking between the
two types of models.
3Engle and Hendry (1989) also apply a formal testing procedure to test for the significance of
the Lucas critique in the case of the demand for Ml in the UK. Their testing procedure is dis-
cussed in detail in Section 2 below. Overall they find little evidence in favour of the relevance of
the critique for Ml.
543

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT