Estimating the elasticity of supply of housing space rather than units

Microeconomic theory suggests that consumers do not consume and suppliers do not supply housing units. Housing units are an artifact of the way in which individuals cluster in residential real estate. The most fundamental measure of willingness to pay on the part of consumers and cost of production experienced by producers is interior space. Empirical literature on housing supply in cities has been based on counts of housing units primarily because counts of units are readily available, across time and space. This paper shows how to produce estimates of aggregate square feet of housing. A theory section then demonstrates that measuring supply elasticities using counts of units are biased downward compared to results using interior space. Finally empirical tests confirm the theoretical evidence that housing supply estimates based on counts of units are too small.