Dissecting highway construction cost inflation: A two-decade analysis of inflation types using the national highway construction cost index

Highway construction cost inflation has reached historic highs, outpaced overall inflation, and significantly eroded the purchasing power of planned funding. Little empirical research has differentiated inflation types, i.e., demand-pull, cost-push, and built-in inflations. Understanding these distinctions is vital for targeted mitigation; for example, addressing input cost increases alone cannot ease demand-driven pressures. Using the National Highway Construction Cost Index (NHCCI) (2003-2023), this study applies various statistical methods - including the Zivot-Andrews, bidirectional Granger causality, cointegration via autoregressive distributed lag bounds, vector error correction model (VECM), out-of-sample forecasting performance, and model diagnostic - to examine short- and long-term relationships between the NHCCI and the potential leading indicators. The results show that in a predictive sense, a wide range of cost-push variables such as producer price indices (PPIs) for materials, equipment, and construction labor productivity, along with certain demand-pull and built-in variables, influence NHCCI, with some exhibiting bidirectional interactions. The results show that cost-push variables (e.g., PPIs for fabricated structural metal, construction materials, construction labor productivity), demand-pull variables (e.g., construction backlog, building permit, housing start), and built-in variables (e.g., M2 money supply, S&P 500 index) are cointegrated with the NHCCI. Following a shock, the NHCCI adjusts toward long-term equilibrium with cost-push variables (i.e., PPIs for fabricated structural metal, construction materials), instead of drifting away. The contributions of this study include (1) dissecting national highway construction cost inflation drawing on established inflation-type theories; (2) identifying PPIs for common line items that outperform traditional high-level macroeconomic variables used in literature; (3) examining bidirectional relationships and feedback mechanisms between the NHCCI and underlying inflation types; and (4) investigating the cointegration between the NHCCI and underlying inflation types. Overall, the study enhances the understanding of the issue and offers practical insights for policymakers, cost estimators, and infrastructure planners in estimating and mitigating highway construction cost inflation.