from the New York Fed
Over 80 percent of Americans live in metropolitan areas, and housing is the dominant land use in cities. For many Americans, homeownership is an important goal, and a substantial majority of the population, including renters, believes that homeownership is a good way to improve their financial situation (Fannie Mae, 2013). For many owners, and for the great majority of renters, purchasing a home will mean obtaining a mortgage. For this reason, the availability and form of mortgages is an important determinant of the homeownership rate, which in turn affects the nature of the housing stock and the organization of residential activity within and across metropolitan areas.
In this chapter, we consider the literature on mortgage finance in the U.S. and its role in shaping the urban landscape. The 2000s witnessed an enormous boom/bust cycle in the residential real estate market, followed by the sharpest contraction in the overall economy since the 1930s. These events, which are widely thought to have been driven at least in part by the mortgage market, had a pronounced spatial pattern which research is only beginning to completely understand. Our workhorse models of local demand and supply of owner-occupied housing can give us only partially satisfactory explanations for the patterns we observe in the data, and more work, both theoretical and empirical, is needed to understand why the boom/bust cycle occurred when and where it did. For example, the user cost framework that has long served as the basis for analyzing how credit conditions affect the demand for owner-occupied housing provides a good basis for understanding the direction of demand shifts during the boom, but the basic model requires significant extensions to capture the magnitude and locational patterns of the demand volatility we observed during the 2000s.
[click on image below to continue reading]
Source: http://www.newyorkfed.org/research/staff_reports/sr713.pdf