AN ECONOMETRIC ANALYSIS OF RETAIL RENTS IN SINGAPORE

Abstract

Primarily based on Alonso’s bid rent model, this study examines the dynamics of the Singapore’s overall retail rental market by adopting a Vector Error Correction Model (VECM) estimation. The proxy used for overall retail rental value is indicated by a combination of the shop rent index from 2004 to 2013 and the retail rent index in 2014, maintained by Singapore’s planning authority, the Urban Redevelopment Authority (URA). The independent variables used in the study are the real gross domestic product, monthly earnings of individuals and vacancy rates. The variables LOGGDP and VR would be cointegrated of order one, I(1), and variables LOGME and LOGSRI would be cointegrated of order two, I(2), to enable them to be used in the VECM. The resulting VECM model shows a good fit that allows the error correction term (ecm) and together with the economic, financial and rental variables to jointly explain about 79.2% of the variation in the overall retail rent index. With a positive CoinEq1 coefficient that is positive and statistically significant at 5% level, it would take time for the system to return to its equilibrium once it has been shocked. Another variable that shows significant explanatory relationships with respect to changes in the overall retail rents, includes past rents (index points) in the second order lags [D(LOGSRI(-2))]. The variable [D(LOGGDP(-3))], with a significant t-statistic value at 2.916265, also helps to explain the changes in the overall rents. Additionally, both the first and third differences of the lagged macroeconomic variables of Monthly Earnings of Individuals are moderately significant. Of the remaining variable, Vacancy Rate, both the first and second differences are significant in accounting for the variation in changes of overall retail rents with their t-statistics values being above 3.0. Lastly, the results from the ex post forecasting estimates show that the VECM is efficient in predicting overall retail rents in Singapore.