Resumen
The majority of policymakers believe that investments in construction infrastructure would boost the economy of the United States (U.S.). They also assume that construction investment in infrastructure has similar impact on the economies of different U.S. states. In contrast, there have been studies showing the negative impact of construction activities on the economy. However, there has not been any research attempt to empirically test the temporal relationships between construction investment and economic growth in the U.S. states, to determine the longitudinal impact of construction investment on the economy of each state. The objective of this study is to investigate whether Construction Value Added (CVA) is the leading (or lagging) indicator of real Gross Domestic Product (real GDP) for every individual state of the U.S. using empirical time series tests. The results of Granger causality tests showed that CVA is a leading indicator of state real GDP in 18 states and the District of Columbia; real GDP is a leading indicator of CVA in 10 states and the District of Columbia. There is a bidirectional relationship between CVA and real GDP in 5 states and the District of Columbia. In 8 states and the District of Columbia, not only do CVA and real GDP have leading/lagging relationships, but they are also cointegrated. These results highlight the important role of the construction industry in these states. The results also show that leading (or lagging) lengths vary for different states. The results of the comparative empirical analysis reject the hypothesis that CVA is a leading indicator of real GDP in the states with the highest shares of construction in the real GDP. The findings of this research contribute to the state of knowledge by quantifying the temporal relationships between construction investment and economic growth in the U.S. states. It is expected that the results help policymakers better understand the impact of construction investment on the economic growth in various U.S. states.