Demand for housing, supply of housing, and inter-city house price movements in the United States

This is an empirical study about the demand for housing, supply of housing, and inter-city house price movements in the United States. In Chapter 1, the relationships between house prices, land prices, construction costs, and construction activity are explored. Apart from an extensive survey of the previous literature, causal relationships are investigated through interviews, regressions, and causality tests. It is argued that house price movements are closely correlated with movements in land prices and that relationships between construction costs and house price movements might be due to feedback from housing demand rather than exogenous changes in construction costs driving movements in house prices. In Chapter 2, household demand for housing services is estimated using panel data for the United States from 1970 to 1987. Housing consumption is shown to depend on a number of demographic, economic, and socio-economic variables, such as the age composition within the household (following Mankiw and Weil, 1989), permanent income, user cost, marital status, education, and race. Results obtained when focusing on homeownership and race are compatible with the existence of discrimination in the real estate and/or mortgage markets. Estimations that are limited to a subsample of recent movers best explain inter-household variations in housing consumption. Chapter 3 presents a model based on DiPasquale and Wheaton (1992) which emphasizes the role of land as an input factor for housing construction and incorporates sluggish adjustment of house prices. A regression equation based on fundamental supply and demand factors is obtained by solving the market clearing condition for the total housing market. The model is estimated for 23 MSAs 1977-1991 employing panel data estimation methods and using the Freddie Mac repeat sales price index and new data on MSA housing units is created by combining annual construction permit data with decennial counts of housing units. Real house price movements are found to be negatively influenced by changes in the total number of housing units per capita and positively influenced by changes in per capita income and by house price appreciation the previous period. The strength of these relationships vary over time and between regions.