Increasing accessibility by public transport benefits local economy: the effect of a new metro line in Rome

TL;DR

The study investigates whether new public transit infrastructure can boost the local economy beyond mobility gains by examining Rome's Metro C expansion. It develops a tailored multiplicative difference-in-differences approach to link accessibility improvements to changes in the number and composition of economic activities and to micro-enterprises. Results show a statistically significant increase in the number of economic activities near Metro C (about $5 \pm 4\%$ by 2018, $p<0.01$) and a mild diversification decline, suggesting clustering toward similar sectors; micro-enterprises also exhibit higher employment and GDP. These findings support the idea that public transport investments promoting sustainable mobility can generate positive spillovers for the local economy, while noting potential heterogeneity and gentrification caveats.

Abstract

This study investigates the economic impact of Metro C, a major expansion of Rome's metro system. Using a difference-in-differences (DID) approach within a multiplicative framework, the research quantifies the impact of increased accessibility on local economic activities. The results show a statistically significant rise in the number of economic activities in areas affected by the new line. A mild decline in economic diversity suggests the emergence of spatial clustering of similar activities. A dedicated analysis of microenterprises, which represent the majority of the dataset, examines changes in employment and GDP associated with the new infrastructure. The observed zone-level correlation between accessibility gains and growth in economic activities also offers a basis for generalising the findings beyond the specific case of Metro C. Overall, the case study shows that public transport investments aimed at boosting sustainable mobility can also generate positive spillover effects on the local economic fabric.

Figures (4)

• Figure 1: Accessibility increase of urban areas and selected treated group. In panel (a), the effect of the introduction of Metro C is measured as the average variation of local accessibility (the number of people who can reach a point using public transport during a typical day) for each zone. Panel (b) shows the density of economic activities and, highlighted, the treated zones. We group the urban zones of Rome in 5 classes based on their density of economic activities in 2013. Classes are marked by shades of blue, Metro C stations are marked with blue dots and treated urban zones are highlighted in yellow. Zones excluded from the study are filled in black.
• Figure 2: DID estimation of the effect of accessibility increase on the number of activities in treated areas. Panel (a) shows the enhancement in the economy of urban zones of Rome interested by the opening of a new metro line with respect to the control group: relative difference in the average number of economic activities per urban zone between zones interested by a new metro opening and zones not touched by it, as a function of the number of years after the openings, for each class of density of activities. Class 4 (the areas with greatest density of economic activities) seems to be the most affected. Transparent curves represent each individual urban zone. In panel (b), global absolute number of economic activities affected by the Metro C opening, computed as the sum of the residuals of each urban zone affected by the new metro with respect to its appropriate baseline. The shadowed region spans a one standard deviation confidence interval. In panel (c), the relationship between the number variation of economic activities due to the introduction of Metro C and the accessibility increase brought by it, for urban zones of Rome. Each dot represents an urban zone, the blue line is a kernel non-parametric regression, the shadowed region corresponds to a $95 \%$ confidence interval on the regression, and treated zones occupy the region of the plot on the right of the dashed grey line, while control zones are on the left. The accessibility increase is quantified as the increase in the number of people reachable by public transport from each area due to the new metro line. In panel (d), the mean relative variation in entropy per urban zone is due to the opening of line C of the metro. Transparent curves represent each individual urban zone.
• Figure 3: Spatial variations in the number and diversity of firms. Panel (a) shows the variation in the number of economic activities across urban zones in Rome in 2018, relative to the baseline trend. Panel (b) displays the variation in the diversity of economic activities, also relative to the baseline trend. Diversity is quantified using Shannon entropy.
• Figure 4: Effects of the introduction of Metro C on micro-enterprises. In (a), temporal evolution of the number of jobs in micro-enterprises attracted in treated zones by the introduction of Metro C. In (b), temporal evolution of the variation in GDP in micro-enterprises due to the introduction of Metro C, in treated zones. The shadowed regions correspond to one standard deviation.