Housing policy outcomes in global perspective: An application of discriminant analysis: Fid-Bau Portal

Housing policy outcomes in global perspective: An application of discriminant analysis

Arimah, Ben C.

Abstract Comparative analysis of the functioning of the housing sector has been severely limited by the dearth of reliable data. In seeking to rectify this, the UNCHS and World Bank have provided data on 52 cities through their joint Housing Indicators Program. Drawing on this data set, this paper identifies the housing policy outcomes that are particularly important in distinguishing various groups of cities. This paper also provides a classification of cities on the basis of their respective housing outcomes. The empirical analysis is based on a discriminant analysis. It reveals that infrastructure expenditure per capita is the most important housing policy outcome that discriminates between the various groups of cities. This is followed by floor area per person, housing credit portfolio, percentage of unauthorized housing, and total investment in housing. Furthermore, the analysis indicates that cities which are identical on the basis of their income levels may not necessarily be identical when their housing policy outcomes constitute the criteria for such classification. This is reflected in the fact that 15 of the 52 cities were reclassified on the basis of their housing policy outcomes.