This paper identifies and juxtaposes different approaches to modelling
housing construction. The first two are based on dynamic marginal pri
cing considerations and emphasize adjustment costs and the necessary t
ime to build respectively. Importantly, both these models introduce so
me degree of forward looking behaviour into the house building decisio
n. Both approaches are able to explain aspects of UK housing starts, 1
970-90. Non-nested testing showed that the time to build story is more
powerful in explaining the data set at hand. However, we have not bee
n able to find any evidence in support of a third hypothesis, namely t
hat quantity signals, proxied by turnover in the housing market, influ
ence starts. The start to completion lag, in turn, is found to be affe
cted by the interest rate representing the opportunity cost of delayed
completions.