TESTING RATIONAL EXPECTATIONS AND EFFICIENCY IN THE LONDON METAL EXCHANGE
| DOI | http://doi.org/10.1111/j.1468-0084.1988.mp50001003.x |
| Author | Ronald MacDonald,Mark P. Taylor * |
| Date | 01 February 1988 |
| Published date | 01 February 1988 |
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 50, 1 (1988)
0305-9049 S3.00
VESTING RATIONAL EXPECTATIONS AND
EFFICIENCY IN THE LONDON METAL
EXCHANGE
Ronald MacDonald and Mark P. Taylor*
Over the last ten years or so a substantial number of tests of the Efficient
Market Hypothesis (EMH) have been conducted for a variety of auction
markets. A market is said to be efficient if market prices 'fully reflect' all
currently available information (see Fama, 1970). The EMH is a joint hypo-
thesis, consisting of the propositions that market participants form their
expectations according to the rational expectations hypothesis (REH) and
have some view of the equilibrium process driving excess returns (either risk
neutrality or risk aversion on the part of market participants may be con-
sistent with this second leg of the joint hypothesis - see, for example, Bilson,
1981). Since, by definition, the EMH cannot be satisfied in a fix-price market,
researchers in their tests of the hypothesis have concentrated on flex-price, or
auction markets such as foreign exchange markets, stock markets, bond
markets and commodity markets.1 Whether or not a market obeys the EMH
has important implications for economic policy since agents transacting in
such a market will ensure an optimal allocation of resources (Fama, 1970,
1976).In testing the EMH in auction markets, researchers usually assume risk
neutrality and test for unbiasedness and forecast error orthogonality. The
unbiasedness property, that the current n-period forward rate f should be an
optimal predictor of the future spot rate (s, +n), is a straightforward implica-
tion of the first moment properties of rational expectations. The error-
orthogonality property, that the current forecast error should be uncorrelated
with information on which agents base their expectations (E ( f - s,) II) 0),
can be viewed as an implication of the second moment properties of rational
expectations (i.e. that agents' forecasts should have minimum conditional var-
iance). A powerful test of both these implications of the EMH has been deve-
loped by Sargent (1979) and popularized by Hakkio (1981) and Baillie,
Lippens and McMahon (1983), and may be termed the Bivariate Vector
Autoregressive (BVAR) approach.
*We are grateful to Martyn Hill of the Commodities Research Unit, London, for help in
obtaining some of the data used in this paper, and to an anonymous referee for helpful
comments on a previous version. The usual disclaimer applies. Any views expressed are those
of the authors and are not necessarily those of the Bank of England.
The distinction between flex price and fix price markets is due to Hicks (1974).
41
To continue reading
Request your trial