Journal of Real Estate Finance and Economics, 18:1, 9±23 (1999)
# 1999 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
Why Don't We Know More About Housing Supply?
City Research, Boston, MA
This paper reviews the main themes in the empirical literature on housing supply and outlines suggestions for
future research. Much of the literature has focused on the determinants of new housing supply, particularly the
supply of single family detached homes, and the renovation and repair decisions of homeowners. We have learned
a great deal from the work that has been done but many important puzzles remain. Much of the literature has
focused on aggregate data because there is so little information where the unit of observation is the builder,
investor, or landlord. We need to focus on bringing new data to bear on the decision-making processes of these
important actors to build our understanding of the micro foundations of housing supply.
Key Words: housing supply, housing markets
Virtually every paper written on housing supply begins with some version of the same
sentence: while there is an extensive literature on the demand for housing, far less has been
written about housing supply. Although this statement is clearly true, at this point, there
have actually been a considerable number of papers on housing supply. However, the
empirical evidence on the supply of housing is far less convincing than that on the demand
for housing. Why don't we know more about housing supply?
Because housing is a durable good, housing supply is determined not only by the
production decisions of builders of new units but also by the decisions made by owners of
housing (and their agents) concerning conversion of the existing stock of housing. Owners
can convert two units into one or convert a large single-family Victorian home into several
small apartments. Owners may decide to rehabilitate an existing unit to increase the ¯ow
of housing services being delivered by that unit or decrease maintenance of an existing
unit, decreasing the ¯ow of housing services. These adjustments to the existing stock are
signi®cant. In 1995, expenditures on maintenance, repair, improvements, and alterations
of the housing stock totaled $111.7 billion; expenditures on private construction of new
residential buildings totaled $162.4 billion.
Baer (1986) argues that while new
construction is the main source of additions to the housing stock, the share of additions
to the stock that come from improvements to the existing stock has increased over time,
rising from roughly 10% of total additions during the 1950s and 1960s to about 27%
between 1973 and 1980.
In addition, government policy can have a profound impact on the operation of the
housing market. Rental assistanceÐsuch as vouchers or subsidies to homeowners in the
form of the mortgage interest deductionÐincreases demand for housing services. The
long-run impact on price depends on the supply response determined by the price elasticity