ECONOMETRIC MODELING OF RENTAL HOUSING SUPPLY IN SHENZHEN, CHINA

Abstract

This study investigates the dynamics of rental housing supply in Shenzhen, one of the fastest growing cities in China and globally. In recent years, rapid urbanization has created housing affordability issues in many large cities. This has resulted in a sharp spike in the number of households seeking rental housing. This increase is not met by the supply of rental housing, resulting in rising rents, overcrowding, long commutes, and homelessness. It is hypothesized that rental housing supply is dependent on GDP per capita, unemployment rate, rental growth, rental rate, vacancy rate, and interest rates. Changes in these variables encourage developers and existing property owners to increase the supply of rental housing. The study does not use land use variables just as zoning and building codes because they are not compatible with time series. A generalized least squares regression design is used because of the presence of heteroscedastic errors. The sample comprises Shenzhen's housing annual rental data from 2001 to 2023 (n=23). Data are collected from different websites such as the National Bureau of Statistics and housing management agencies in Shenzhen. Because the data are collected on an annual basis, the model does not include short-term adjustment dynamics, that is, quarterly lags in the independent variables. The main findings are that the supply of rental housing is primarily affected by the unemployment rate (+), rental growth (+), and GDP per capita (?). The positive relationship for unemployment may appear surprising. However, Shenzhen is a magnet for migrants, and a higher unemployment rate indicates that more people may be looking for work and rental housing. The implications are that rental housing supply is responsive to rental rate (rent per month) and rental growth rather than financial variables such as interest rates.