From linear to nonlinear models: Responsiveness of house prices to shocks from macroeconomic indicators in Kenya

Previous studies have concluded that models that account for nonlinear relationships outperform their linear counterparts. However, while this is evident in advanced economies, the same cannot be inferred in Sub-Saharan Africa. This paper seeks to show how house prices are responsive to the nonlinear shocks of selected macroeconomic indicators, while controlling for the effects of interest rate and population in the Kenyan housing market. We employed the Nonlinear Autoregressive Distributed Lag (NARDL) model on a quarterly dataset from 2004Q1 to 2021Q4. In the long run, we showed that house prices respond to some nonlinear shocks of macroeconomic indicators, particularly shocks of the exchange rate and the shock of the market index, but not inflationary rate. Moreover, we established a significant negative effect of interest rate and population on house prices. In the short run, we also noted that a decrease in the exchange rate influences house prices negatively. Interest rate, including its lag terms, also impacts house prices negatively. These findings call for prudent macro-prudential policies, more importantly: (a) management of exchange rate and interest rate; (b) give property investment insight and (c) offer effective policy direction in the development and sustenance of the Kenyan property market.