The Lead-Lag Relationships between Construction Investment and GDP: Granger Causality Tests and Impulse Responses Using Japanese Data

This paper analyzes the lead-lag relationships between construction investment and GDP at business-cycle frequency by the tests of Granger causality and the impulse responses in the framework of the vector autoregression, using the annual Japanese data. The analysis find that private construction investment tends to lead GDP, not vice versa, in the Granger sense and is of value in predicting the course of GDP one year ahead. Government construction investment, on the other hand, tends to lag GDP.