Housing cycles and efficiency

AuthorGeoffrey Meen
Publication Date01 May 2000
Scottish Journal of Political Economy, Vol. 47, No. 2, May 2000
#Scottish Economic Society 2000. Published by Blackwell PublishersLtd, 108 Cowley Road, Oxford OX4 1JF, UK and
350 Main Street, Malden, MA 02148, USA
Geoffrey Meen
This paper has four objectives. First, a small model of the UK housing market is
constructed, including equations for house prices, housing starts, construction costs
and interest rates. The model is used in an analysis of housing market cycles,
employing techniques developed for the analysis of general business cycles. Second,
the model is used to consider housing market efficiency. Third, the model is
extended to examine the relationship between house prices and property
transactions. Finally, the role of monetary policy in the generation of housing
cycles and stability is discussed.
This paper is concerned with two of the more important issues in empirical
housing economics over the last fifty yearsÐ housing cycles (or stability) and
efficiency. It begins by constructing a small, fully estimated quarterly model of
the UK housing market, using techniques taken from the more general business
cycle literature, which provides the basis for the subsequent analysis. It, then,
explores housing market efficiency within the structure of the model. The
efficiency conditions lead on to a recent area of housing researchÐ the
relationship between house prices and property transactions. In an efficient
market, one strand of the literature suggests that there should be no relationship
between the two variables although, in practice, a strong correlation has been
observed, particularly in the US. Finally, the paper returns to an old theme, but
set in a new context. What, if any, is the relationship between the stance of
monetary policy and housing cycles, particularly in the light of government
reliance on explicit inflation targets? Do targets add to or reduce stability in the
housing market?
For the first part, our model initially consists of four equations, covering
housing starts, house prices, construction costs and interest rates which may be
modelled as a VAR or equivalently as a vector error correction model (VECM).
From the VECM we construct a structural model that has a ready economic
interpretation, imposes identification restrictions on the system and is
appropriately conditioned on certain weakly exogenous variables. Furthermore,
University of Reading
the model appears to be fully data coherent and, hence, provides a suitable
vehicle for examining both cycles and long-run trends in the housing market.
In the second part of our study, the house price equation gives rise to a
housing market efficiency condition that has been the subject of a large number
of tests, particularly on US data. Much of this literature suggests that housing
markets are inefficient (see, for example, Gatzlaff and Tirtiroglu, 1995, or Cho,
1996, for surveys). The efficiency conditions are closely related to the housing
user cost of capital. Therefore, under a range of different shocks to the model,
changes in the user cost and associated measures of the excess return to housing
can be used as a test of housing market efficiency. The adjustment processes that
operate within the housing market are critical. For the third part of the paper,
the framework of the model set out earlier is used to conduct the first empirical
tests of the relationship between changes in house prices and transactions in the
Finally, the importance of monetary policy for housing cycles arises from the
nature of the adjustment processes in housing referred to above. We suggest that
the level of interest rates required to meet inflation targets under well-known
policy rules is not necessarily consistent with the level required for housing
market stability. A policy inconsistency arises from the weak supply elasticities
that characterise the UK housing market. Under some circumstances, reliance
on interest rate policy rules may increase the volatility of the housing market,
when subjected to random shocks.
Section II discusses in a little more detail the nature of the user cost of capital
and its relationship to the housing efficiency literature. The model is estimated in
Section III. Section IV discusses the nature of housing market adjustment in
response to shocks, whereas Section V estimates the model of prices and
transactions. In Section VI, the model is subjected to a range of shocks, both
random and policy orientated. These allow us to consider cycles, stability, the
effects of violation of the efficiency conditions and the extent of correlation in
the excess return to housing. The simulations also demonstrate that it is possible
to design a monetary policy that leads to a reasonably stable housing market. In
Section VII, the VECM is embedded in a full econometric model of the UK
economy; repeating the shocks, it can be demonstrated that adherence to interest
rate rules not directly related to the housing market may increase the size of
housing cycles.
The housing user cost and measures of housing market efficiency are closely
related. As is well-known, e.g. Meen (1990), the user cost is derived from the
marginal rate of substitution between housing and a composite consumption
good, given by (1), and measures the real price of owner-occupier services. It
may be extended to include property taxes, maintenance expenditures,
transactions costs and credit constraints.
h=cg(t)[(1 )i(t)_
g=g(t)] (1)
#Scottish Economic Society 2000

To continue reading

Request your trial