Housing Search Activity and Quantiles-Based Predictability of Housing Price Movements in the United States

Recent evidence from a linear econometric framework, tend to suggest that housing search activity, as captured from Google Trends data, can predict housing returns of the overall United States (US), as well as at the regional-level for Metropolitan Statistical Areas (MSAs). Based on search-theory, we, however, postulate that search activity can also predict housing returns volatility. Given this, we use a k-th order nonparametric causality-in-quantiles test, which in turn, allows us to test for predictability in a robust manner over the entire conditional distribution of not only housing price returns, but also its volatility (i.e., squared returns), by controlling for nonlinearity and structural breaks that exists in the data. Using this model, over the monthly period of 2004:01 to 2021:01, we show that while housing search activity continues to predict aggregate US house price returns barring the extreme ends of the conditional distribution, volatility is relatively strongly predicted over the entire quantile range considered. Our results tend to carry over to an alternative (the Generalized Autoregressive Conditional Heteroskedasticity (GARCH)-based) metric of volatility, higher (weekly)-frequency data (over January, 2018-March, 2021), as well as to over 84% of the seventy-seven MSAs considered. Our findings have important implications for investors and policymakers, as well as academics.