In this note, using a four-variable VAR model, we empirically test the causal link between construction activity and economic growth in Taiwan over the 1979Q1–1999Q4 period. The cointegration results of Johansen [J. Econ. Dyn. Contr. 12 (1988) 231] and Johansen and Juselius [Oxford Bull. Econ. Stat. 52 (1990) 169] indicate that there exists a cointegrating vector among four variables—namely, real GDP, real investment in construction activity, real government expenditures, and real private consumption expenditures. Here, the results from Granger-causality tests based on corresponding vector error-correction models (ECM) clearly point to unidirectional causality running from construction activity to economic growth. This, the major finding of this study, has important implications with respect to developing economic policy regarding the role of construction activity in Taiwan over the test period.