Abstract
<title>Abstract</title> <p>Most empirical evidence on road toll elasticities is derived from studies of toll implementation or toll increases and typically relies on aggregate traffic or aggregated truck flows. Consequently, relatively little is known about how truck demand responds when tolls are removed, even though toll removal represents a clean pricing experiment that avoids the confounding effects often associated with new toll facilities or infrastructure upgrades. Moreover, existing studies rarely distinguish between different segments of truck traffic, despite the critical role heavy vehicles play in toll financing by contributing a disproportionately large share of toll revenues. This study addresses these gaps by examining how light and heavy trucks respond to road toll removal using revealed-preference data. The analysis draws on traffic observations from 14 toll road projects in Norway and focuses exclusively on road segments where tolls were removed without any accompanying changes in infrastructure. This multi-project time-series setting allows demand responses to be identified both over time and across projects. An autoregressive distributed lag (ARDL) model combined with an unrestricted error correction model (UECM) is employed to capture short-run dynamics and long-run equilibrium effects. The results reveal clear and systematic differences between light and heavy trucks. Short-run elasticities for light trucks typically range between − 0.2 and − 0.8 and are statistically significant in several cases, indicating an immediate response to toll removal. Long-run elasticities for light trucks are significant in fewer projects and typically lie between − 0.3 and − 1.0, suggesting that persistent demand adjustments occur only in a subset of cases. In contrast, heavy truck elasticities are substantially smaller, averaging around − 0.1 and frequently statistically insignificant in both the short run and the long run, indicating limited responsiveness to toll removal. Overall, the findings suggest that traffic responses to toll removal are primarily short-lived and concentrated among light trucks, while heavy truck demand remains largely insensitive to pricing changes. These results highlight the importance of distinguishing between truck categories in toll appraisal, revenue forecasting, and transport modeling, and suggest that toll revenues derived from heavy vehicles may be more stable over time than commonly assumed.</p>