Temporal causality between house prices and output in the US : a bootstrap rolling-window approach

Abstract

This paper examines the causal relationships between the realhouse price index and real GDP per capita in the US, using thebootstrap Granger (temporal) non-causality test and a fixed-sizerolling-window estimation approach. We use quarterly time-seriesdata on the real house price index and real GDP per capita, cov-ering the period 1963:Q1 to 2012:Q2. The full-sample bootstrapnon-Granger causality test result suggests the existence of a uni-directional causality running from the real house price index toreal GDP per capita. A wide variety of tests of parameter constancyused to examine the stability of the estimated vector autoregres-sive models indicate short- and long-run instability. This suggeststhat we cannot rely on the full-sample causality tests and, hence, this warrants a time-varying (bootstrap) rolling-window approachto examine the causal relationship between these two variables.Using a rolling window size of 28 quarters, we find that whilecausality from the real house price to real GDP per capita occursfrequently, significant, but less frequent, evidence of real GDP percapita causing the real house price also occurs. These results implythat while the real house price leads real GDP per capita, in general(both during expansions and recessions), significant feedbacks alsoexist from real GDP per capita to the real house price.