ITEA Conference and ITEA Annual School on Transportation Economics

June 17, 2026
Time	Location	Event
12:30 to 13:30	Palazzo Baroni Roof	Registration and Lunch
13:30 to 15:00		Parallel Sessions 1, Palazzo Baroni (Pignolo)
15:30 to 17:00		Parallel Sessions 2, Palazzo Baroni (Pignolo)
17:00 to 17:15	Palazzo Baroni 17 and Garden	Coffee Corner
17:15 to 18:15		Parallel Sessions 3, Palazzo Baroni (Pignolo)
18:30 to 20:30	Palazzo Baroni 17 and Garden	Welcome Reception
June 18, 2026
Time	Location	Event
08:30 to 09:00	Palazzo Baroni 17 and Garden	Coffee Corner
09:00 to 10:30		Parallel Sessions 4, Palazzo Baroni (Pignolo)
10:30 to 11:00	Palazzo Baroni 17 and Garden	Coffee Corner
11:00 to 12:15	Aula Magna	Keynote Lecture
12:15 to 13:30	Palazzo Baroni 17 and Garden	Lunch
13:30 to 15:00		Parallel Sessions 5, Palazzo Baroni (Pignolo)
15:00 to 15:30	Palazzo Baroni 17 and Garden	Coffee Corner
15:30 to 16:30		Parallel Sessions 6, Palazzo Baroni (Pignolo)
17:00 to 18:45	Città Alta	Guided Tour
20:00 to 23:00	Ristorante Il Pianone	Social Dinner, Ristorante Il Pianone
June 19, 2026
Time	Location	Event
08:30 to 09:00	Palazzo Baroni 17 and Garden	Coffee Corner
09:00 to 10:30		Parallel Sessions 7, Palazzo Baroni (Pignolo)
10:30 to 11:00	Palazzo Baroni 17 and Garden	Coffee Corner
11:00 to 12:00	Palazzo Baroni 16	Awards and Celebration of Honorary Membership
12:00 to 13:15	Palazzo Baroni 17 and Garden	Lunch
13:15 to 14:15		Parallel Sessions 8, Palazzo Baroni (Pignolo)
14:15 to 14:30	Palazzo Baroni 17 and Garden	Coffee Corner
14:30 to 15:30		Parallel Sessions 9, Palazzo Baroni (Pignolo)
15:30 to 16:00	Palazzo Baroni 17 and Garden	Coffee Corner
16:00 to 17:00		Parallel Sessions 10, Palazzo Baroni (Pignolo)
17:15 to 17:45	Palazzo Baroni 16	ITEA General Assembly
18:00 to 19:00	Palazzo Baroni 16	ITEA Executive Committee Meeting (by invitation only)
19:00 to 22:00	Nazionale 1894	Farewell Party (ticketed event; €50 participation fee), Nazionale 1894

Summary of All Sessions
Click here for an index of all participants

#	Date/Time	Title/Location	Papers
1	June 17, 2026 13:30-15:00	Shipping, Ports, and Landlocked Economies     Location: Palazzo Baroni 8	3
2	June 17, 2026 13:30-15:00	Fare Discounts and Subscriptions     Location: Palazzo Baroni 9	3
3	June 17, 2026 13:30-15:00	Information, Experience, and Public Support for Road Pricing     Location: Palazzo Baroni 11	3
4	June 17, 2026 13:30-15:00	Commuting: Costs, Tax, and Time Allocation     Location: Palazzo Baroni 12	3
5	June 17, 2026 13:30-15:00	Urban Public Transport: Appraisal and Design     Location: Palazzo Baroni 13	3
6	June 17, 2026 13:30-15:00	Slots and Capacity Allocation     Location: Palazzo Baroni 14	3
7	June 17, 2026 15:30-17:00	Parking Markets and Curbside Regulation     Location: Palazzo Baroni 8	3
8	June 17, 2026 15:30-17:00	Transit Theory and Design Principles     Location: Palazzo Baroni 9	3
9	June 17, 2026 15:30-17:00	Road Pricing: Design, Elasticity, and Equity     Location: Palazzo Baroni 11	3
10	June 17, 2026 15:30-17:00	Rail Demand and Modal Choice     Location: Palazzo Baroni 12	3
11	June 17, 2026 15:30-17:00	Aviation Policy & Competition     Location: Palazzo Baroni 13	3
12	June 17, 2026 15:30-17:00	Airline Schedule Competition and Cooperation     Location: Palazzo Baroni 14	3
13	June 17, 2026 17:15-18:15	Air–HSR Competition under Liberalization     Location: Palazzo Baroni 8	2
14	June 17, 2026 17:15-18:15	Autonomous Vehicles in Transit Systems     Location: Palazzo Baroni 9	2
15	June 17, 2026 17:15-18:15	Distributional Effects and Tradable Permits     Location: Palazzo Baroni 11	2
16	June 17, 2026 17:15-18:15	Commuting and Refueling Choices     Location: Palazzo Baroni 12	2
17	June 17, 2026 17:15-18:15	Advanced Air Mobility: Markets and Acceptance     Location: Palazzo Baroni 13	2
18	June 17, 2026 17:15-18:15	Entry and Spatial Concentration in Airline Markets     Location: Palazzo Baroni 14	2
19	June 18, 2026 9:00-10:30	Open-Access Competition and Liberalization in European Rail     Location: Palazzo Baroni 8	3
20	June 18, 2026 9:00-10:30	Free-Fare Transit Policies     Location: Palazzo Baroni 9	3
21	June 18, 2026 9:00-10:30	Spatial Economics of Transport and Mobility     Location: Palazzo Baroni 10	3
22	June 18, 2026 9:00-10:30	Managing Road Capacity: Lanes, Tolls, and Governance     Location: Palazzo Baroni 11	3
23	June 18, 2026 9:00-10:30	Urban Air Quality and Vehicle Restrictions     Location: Palazzo Baroni 12	3
24	June 18, 2026 9:00-10:30	Parking and Cruising: In Honor of Donald Shoup     Location: Palazzo Baroni 13	3
25	June 18, 2026 9:00-10:30	Aviation Emissions: Policy and Regulation     Location: Palazzo Baroni 14	3
26	June 18, 2026 11:00-12:15	Keynote Lecture: Joan Walker     Location: Aula Magna	1
27	June 18, 2026 13:30-15:00	Rail Investment, Capacity, and Welfare     Location: Palazzo Baroni 8	3
28	June 18, 2026 13:30-15:00	Transport Appraisal, Planning, and Efficiency     Location: Palazzo Baroni 9	3
29	June 18, 2026 13:30-15:00	Active Mobility: Cycling, Bike Sharing, and Infrastructure     Location: Palazzo Baroni 10	2
30	June 18, 2026 13:30-15:00	Behavioral Responses to Road and Fuel Prices     Location: Palazzo Baroni 11	3
31	June 18, 2026 13:30-15:00	Value of Time and Demand Model Identification     Location: Palazzo Baroni 12	3
32	June 18, 2026 13:30-15:00	Air–Rail Connectivity and Passenger Choice     Location: Palazzo Baroni 13	3
33	June 18, 2026 13:30-15:00	Airline Profitability and Performance     Location: Palazzo Baroni 14	3
34	June 18, 2026 15:30-16:30	Freight Carriers: Market Structure and Pricing     Location: Palazzo Baroni 8	2
35	June 18, 2026 15:30-16:30	Public Transit in Developing-Country Contexts     Location: Palazzo Baroni 9	2
36	June 18, 2026 15:30-16:30	Experimental Evidence on Toll Acceptability     Location: Palazzo Baroni 11	2
37	June 18, 2026 15:30-16:30	Information Frictions in Vehicle Use     Location: Palazzo Baroni 12	2
38	June 18, 2026 15:30-16:30	Senior Drivers and License Renewal Policy     Location: Palazzo Baroni 13	2
39	June 18, 2026 15:30-16:30	Fleet and Network Emissions     Location: Palazzo Baroni 14	2
40	June 19, 2026 9:00-10:30	Transit Demand: Behavior and Competition     Location: Palazzo Baroni 8	3
41	June 19, 2026 9:00-10:30	Macroscopic Models of Traffic Flow and Congestion     Location: Palazzo Baroni 9	3
42	June 19, 2026 9:00-10:30	Road Pricing in Spatial General Equilibrium     Location: Palazzo Baroni 11	2
43	June 19, 2026 9:00-10:30	Cost-Benefit Analysis of Infrastructure: Biases and Distribution     Location: Palazzo Baroni 12	3
44	June 19, 2026 9:00-10:30	EVs: Tax Reform and the Fuel-Tax Transition     Location: Palazzo Baroni 13	3
45	June 19, 2026 9:00-10:30	Airline Operations, Delays, and Conflict     Location: Palazzo Baroni 14	3
46	June 19, 2026 11:00-12:00	Paper Awards and Celebrating Sergio Jara-Diaz as the 2026 ITEA Honorary Member     Location: Palazzo Baroni 16	0
47	June 19, 2026 13:15-14:15	Autonomous Vehicles: Deployment, Accessibility, and Adoption     Location: Palazzo Baroni 8	2
48	June 19, 2026 13:15-14:15	Urban Transport Infrastructure and Welfare     Location: Palazzo Baroni 9	2
49	June 19, 2026 13:15-14:15	Modal Substitution and Multimodal Competition     Location: Palazzo Baroni 11	2
50	June 19, 2026 13:15-14:15	Rural and Urban Mobility: Choices and Consequences     Location: Palazzo Baroni 12	2
51	June 19, 2026 13:15-14:15	EVs: Adoption Incentives and Subsidies     Location: Palazzo Baroni 13	2
52	June 19, 2026 13:15-14:15	Market Power in US Airlines     Location: Palazzo Baroni 14	2
53	June 19, 2026 14:30-15:30	Pricing and Dynamic Equilibrium     Location: Palazzo Baroni 8	2
54	June 19, 2026 14:30-15:30	Traffic Models, Networks & Simulation     Location: Palazzo Baroni 9	2
55	June 19, 2026 14:30-15:30	Motorways: Causal Effects & Resilience     Location: Palazzo Baroni 11	2
56	June 19, 2026 14:30-15:30	Transport Affordability     Location: Palazzo Baroni 12	2
57	June 19, 2026 14:30-15:30	EVs: Consumer Behavior and Use     Location: Palazzo Baroni 13	2
58	June 19, 2026 14:30-15:30	Cabotage and Dynamic Capacity     Location: Palazzo Baroni 14	2
59	June 19, 2026 16:00-17:00	Structural Spatial Models     Location: Palazzo Baroni 8	2
60	June 19, 2026 16:00-17:00	Transit Service Design and Competition     Location: Palazzo Baroni 9	2
61	June 19, 2026 16:00-17:00	European Rail: Regional and Cross-Border     Location: Palazzo Baroni 11	2
62	June 19, 2026 16:00-17:00	Fuel and Energy Pricing: Welfare Effects     Location: Palazzo Baroni 12	2
63	June 19, 2026 16:00-17:00	EVs: Charging Markets and Infrastructure     Location: Palazzo Baroni 13	2
64	June 19, 2026 16:00-17:00	Airport Performance and Gates     Location: Palazzo Baroni 14	2

 

64 sessions, 155 papers, and 0 presentations with no associated papers

ITEA Conference and ITEA Annual School on Transportation Economics
Detailed List of Sessions

Session 1: Shipping, Ports, and Landlocked Economies June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 8
Session Chair: Gernot Sieg, University of Münster
Landlockedness as development problem, when Abstract There are differences in economic performance across countries. Despite most studies have documented the negative effect of being landlocked as one of the determinants for slower economic growth, the methods and data employed by previous studies are not strong enough. Hence, this study will re-examine the association between being landlocked and economic growth. The study will employ an augmented neoclassical economic growth model with panel datasets and correlated random effects estimators. The study will contribute towards the creation of a clear understanding concerning the impact of landlocked on economic growth, through contesting the existing literature.
By Tesfaye Takele; HiMolde
Presented by: Tesfaye Takele, HiMolde
Regulatory Shocks and Contract Design in Port Concessions: A Principal–Agent Approach under the EU ETS Abstract This paper develops a simple principal–agent model to study optimal contract design in port concessions under regulatory uncertainty. The analysis focuses on landlord port systems, where the port authority (principal) grants a long-term concession to a private terminal operator (agent). The operator exerts unobservable effort that affects terminal performance (e.g., productivity, service quality, commercial activity), while demand is subject to exogenous shocks. The introduction of the EU ETS is modeled as an additional regulatory shock that affects traffic volumes independently of operator effort.
By Soraya Hidalgo-Gallego; Universidad de Cantabria    Ramón Núñez-Sánchez; Universidad de Cantabria
Presented by: Ramón Núñez-Sánchez, Universidad de Cantabria
Climate Stress and Strategic Choice: Market Impacts of Droughts in Inland Waterway Transport Abstract Inland waterway transport on the Rhine exhibits a persistent dual market structure characterized by the coexistence of owner-operators and multi-vessel shipping companies. While hierarchical firms benefit from economies of scale in financing, compliance, and coordination, they do not dominate market shares. This paper argues that hydrological uncertainty provides an explanation for this puzzle. We develop a two-state model in which drought conditions reduce vessel capacity and increase spot market prices. Shipping companies operate under long-term contracts and must rely on costly spot market capacity during low-water events. Anticipating this exposure, firms optimally adjust fleet size ex ante. We show that drought-induced navigability risk can offset scale advantages and lead to smaller optimal fleets, particularly in the liquid bulk segment where willingness to pay rises sharply during droughts. The model provides a structural explanation for persistent organizational heterogeneity in the Rhine inland waterway market.
By Till Fischer; University of Münster    Gernot Sieg; University of Münster
Presented by: Gernot Sieg, University of Münster
Session 2: Fare Discounts and Subscriptions June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 9
Session Chair: Anna Matas, Universitat Autonoma de Barcelona
An optimal Deutschlandticket Abstract This paper develops a welfare based model to determine the optimal pricing structure of the Deutschlandticket, a nationwide monthly public transport subscription that provides unlimited access to local and regional services across Germany. The analysis explicitly considers the joint optimization of the uniform national subscription fee and spatially differentiated single fares within a unified welfare framework. The model integrates heterogeneous geographical contexts such as rural areas, cities and large metropolitan regions, incorporates multiple transport modes, and endogenously captures congestion and crowding effects generated by tariff induced demand changes.Social welfare is defined as the sum of consumer surplus and operator profits net of external transport costs. Preliminary results suggest an optimal subscription price above current levels, while optimal single fares remain moderate across areas.
By Daniele De Santis; Sapienza University of Rome    Alejandro Tirachini
Presented by: Daniele De Santis, Sapienza University of Rome
Effects of massive fare discounts on bus ridership and welfare: Evidence from Barcelona Abstract In September of 2022, significant discounts on public transport fares were implemented across Spain. Specifically in Barcelona, monthly passes (which became the most used ticket) received a 50% discount. With data of validations on bus services in the Province of Barcelona, we study the demand effects of the policy through a Fixed Effects model and a Regression Discontinuity in Time analysis, accounting for the heterogeneity of bus services. We obtain an overall increase in ridership of 5% and a price elasticity of -0.058, much less than those reported in the literature. We observe higher ridership response in those services with less transit alternatives and higher level of service. On the other hand, given this low ridership effect and the low cross-elasticities between transit fare and car trips, we estimate a cost-benefit ratio of nearly 1, suggesting that other policies with higher social profitability should be prioritized.
By Albert Gragera; Universitat Autònoma de Barcelona    Anna Matas; Universitat Autonoma de Barcelona    Juan Pablo Sepúlveda Celis; Universitat Autònoma de Barcelona
Presented by: Juan Pablo Sepúlveda Celis, Universitat Autònoma de Barcelona
Are fare discounts effective? Exploring its impact on transit ridership and car substitution Abstract This paper analyses the causal impact of large-scale public transport fare discounts on transit ridership and modal substitution, focusing on the scheme implemented in Barcelona and its metropolitan area from September 2022 onwards. The Spanish government introduced a free short-distance rail pass operated by RENFE and funded a 30% discount on multi-modal tickets, which the metropolitan authority further expanded by an additional 20%, with the objective of mitigating inflationary pressures and encouraging a shift away from private car use Using daily station-level ridership and metropolitan traffic count data for 2018–2024, we combine fixed effects models with regression discontinuity designs in time, exploiting the exogenous policy implementation date. We flexibly control for seasonality, COVID-19 restrictions, price changes, and other concurrent policies. Results from both fixed effects models and regression discontinuity designs consistently point to an 8–10% increase in transit ridership following the fare reductions. Estimated fare elasticities range between –0.17 and –0.29, in line with existing literature, and effects are relatively homogeneous across corridors and fare zones. In contrast, the impact on road traffic is negligible. Most specifications show no statistically significant reduction in car traffic; where significant, effects are extremely small, with a cross-elasticity around –0.01. Overall, our findings suggest that although fare discounts increase public transport demand, their capacity to induce substantial modal shift from private cars is limited, raising questions about their cost-effectiveness as a standalone policy for reducing car dependency and improving welfare.
By Albert Gragera; Universitat Autònoma de Barcelona    Anna Matas; Universitat Autonoma de Barcelona    Jordi Perdiguero; Universitat Autonoma de Barecelona
Presented by: Anna Matas, Universitat Autonoma de Barcelona
Session 3: Information, Experience, and Public Support for Road Pricing June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 11
Session Chair: Jonathan Hall, University of Alabama
Public Support for Universal Road Pricing Abstract Distance-based road pricing can internalize driving externalities and provide fiscally sustainable infrastructure financing as vehicle electrification erodes fuel tax revenues. Yet, political resistance may pose a barrier to implementation. We provide the first comprehensive analysis of preferences for universal road pricing using a stated choice experiment with nearly 4,000 Norwegian respondents. Policy alternatives vary systematically across urban peak prices, off-peak prices, non-urban prices, electric vehicle prices, and revenue allocation. We document robust majority support (70% to 80%) for most universal road pricing schemes, substantially exceeding typical support for local congestion charges. Support varies by design: public transport earmarking outperforms general budget allocation by 8 percentage points, while moderate pricing dominates high pricing by 5 to 7 points. Support varies substantially with car use, availability of transport alternatives, and institutional trust, but shows little sensitivity to income, urban residence, or toll exposure. Latent class analysis identifies unconditional supporters (57%), systematic opponents (22%), and a persuadable middle (21%). Randomized information treatments generate modest effects, with factual information increasing support by 6 percentage points. Political barriers to road pricing may be lower than assumed with moderate pricing and credible infrastructure earmarking
By Alberto Antonioni; Carlos III University of Madrid    Alice Ciccone; Institute of Transport Economics    Cloe Garnache; Oslo Metropolitan University    Sara Ghivarello; Carlos III University    Askill H. Halse; Institute of Transport Economics    Francesca Lipari; LUMSA University (LIBERA UNIVERSITÀ MARIA SS. ASSUNTA)
Presented by: Askill H. Halse, Institute of Transport Economics
Does Participating in a Tolling Experiment Increase Support for Urban Road Pricing? Abstract Congestion pricing is widely regarded as an efficient instrument to address urban traffic externalities, yet it often faces substantial political resistance prior to implementation. Several real-world cases (including Stockholm, Milan, and London) show that public support tends to increase after implementation. This pattern is commonly attributed to learning from direct policy experience. However, real-world implementation combines exposure to toll payments with observable reductions in congestion, improved travel times, and environmental benefits, making it difficult to identify the underlying mechanism. This paper isolates the private monetary channel. We conduct a controlled field experiment with approximately 250 urban drivers in Austria. Participants are recruited via a smartphone app that tracks driving behavior, allowing for a ten-week pre-measurement period. During a three-week intervention phase, drivers face a €7 charge when entering or exiting designated central areas (maximum once per day). Each participant receives an upfront travel budget; toll payments are deducted and any remaining balance is paid out. Support for introducing a city toll in Vienna is measured before and after the intervention. We find no aggregate change in support. Although 51% of participants revise their position, revisions are symmetric and cancel out. Baseline attitudes strongly predict post-experimental support, and realized budget balances have no robust effect. The results suggest that direct cost experience alone does not increase acceptability; post-implementation gains in support likely depend on the realization of collective benefits rather than on exposure to toll payments per se.
By Stefanie Peer; Vienna University of Economics & Business
Presented by: Stefanie Peer, Vienna University of Economics & Business
The New York Congestion Charge Abstract Traffic congestion is one of the most important issues in cities, where most people live. But congestion charges are rarely implemented -- mostly because ex-ante unpopular. This paper examines the New York congestion charge. It focuses on policy evaluation and the causal analysis of public support. Policy evaluation relies on both administrative data and unique survey data. We also examine the extent to which information about and direct experience with congestion pricing are substitutes or complements. We find that congestion pricing is effective at reducing traffic, which leads to many other co-benefits. Public support increases with experience, including outside of the New York metropolitan area. Perceptions about congestion pricing change accordingly. The effect of information is also important, but short-lived.
By Jonathan Hall; University of Alabama
Presented by: Jonathan Hall, University of Alabama
Session 4: Commuting: Costs, Tax, and Time Allocation June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 12
Session Chair: Jos van Ommeren, Vrije Universiteit Amsterdam
Time Allocation and Mode Choice under Flexible Working Regimes: A Joint Continuous-Discrete Model with GPS-Tracking Data from Germany Abstract The post-pandemic rise of remote work has altered commuting patterns and may reshape how individuals value travel time and leisure. Yet existing applications of the Jara-Diaz time allocation framework – which jointly derives the value of leisure (VoL) and mode-specific values of travel time savings (VTTS) – have not accounted for remote work. This paper addresses this gap using GPS-tracking data from 473 employed individuals in Germany, collected during May 2023 as part of a larger mobility panel. We estimate a joint continuous-discrete model combining a Cobb-Douglas time allocation block with a multinomial logit mode choice model. Individual hourly wages are computed from reported income brackets and GPS-observed working time, and detailed trip purposes are exploited through interaction terms. A preliminary standalone logit model suggests that each additional home office day per week reduces the implied VTTS for car travel from 56 to 51 EUR/h. Descriptively, frequent home office users show substantially lower car shares and higher walking shares on commute trips. These findings contrast with stated preference evidence from the pandemic period and suggest that post-pandemic revealed preferences may reflect a more stable behavioral equilibrium. Full joint model results, including VoL and purpose-specific VTTS segmented by home office intensity, will be presented at the conference.
By Florian Dix; DIW Berlin    Heike Link; DIW
Presented by: Florian Dix, DIW Berlin
The Cost of Commuting in Germany: Combining Expenditure and Travel Diary Data before and after the COVID-19 pandemic Abstract Commuting trips between home and work account for around 15% of all trips in Germany. While commuting distances have increased over the past decades, post-COVID remote work reduced commuting volumes. Despite extensive international evidence on commuting, with commuting costs being a central determinant of mode choice, commuting costs have not yet been quantified for Germany in a representative, statistically robust way that captures both direct and indirect costs. This paper estimates generalized commuting costs and relates them to observed commuting behavior. We combine the German Income and Expenditure Survey (EVS) 2018, which provides transport-related spending by cost component, with the national travel survey Mobility in Germany (MiD) 2017 and 2023, which provides trip diaries but limited expenditure information. We estimate direct cost components in EVS and predict them for MiD households using harmonized socio-demographics, spatial, and ownership covariates. We compare econometric approaches (Two-Part and Tobit) with machine-learning methods (Gradient Boosting), and allocate predicted direct costs to commuting based on commuting distance shares. Indirect costs are measured by monetizing commuting durations using net household income as a proxy for the Value of Time. To extend EVS-based costs to 2023, we update cost components using official price indices. First descriptive evidence shows that mean commuting distance slightly decreased from 16.0 km to 15.4 km, while commuting duration increased from 28.6 to 29.1 minutes (2017 vs. 2023), implying potentially higher time-related costs despite shorter distances. Model performance differs by cost component, favoring a machine-learning two-step approach.
By Nicole Reinfeld; Frankfurt University of Applied Sciences    Tobias Hagen; Frankfurt University of Applied Sciences
Presented by: Nicole Reinfeld, Frankfurt University of Applied Sciences
Tax Avoidance, Rich Workers, and Employer-Provided Cars Abstract High marginal income tax rates create incentives for high-income workers to substitute taxable wage income with tax-favoured fringe benefits. In many European countries, employer-provided cars are the main discretionary fringe benefit and are taxed based on administratively assessed values that are typically below their market value. This paper examines whether high-income workers respond to marginal tax rates by adjusting both the probability of receiving a company car (extensive margin) and the value of the car conditional on receipt (intensive margin). Using Swedish administrative panel data covering the period 2005–2022, we exploit variation in marginal tax rates arising from national tax reforms. We estimate fixed-effects and instrumental-variable models, where the instrument is a counterfactual marginal tax rate computed using the car’s market value rather than its administrative value. This approach addresses endogeneity arising from compensation shifting across tax brackets. We find substantial substitution toward employer-provided cars when marginal tax rates increase. A 10 percent increase in the marginal tax rate raises the market value of company cars by approximately 6–9 percent among recipients and increases the probability of receiving a company car by around 2–4 percent. Combining both margins implies a sizable overall elasticity of car compensation with respect to the net-of-tax rate. Our results identify a quantitatively important avoidance channel operating through in-kind compensation. While substitution toward company cars may reduce the effective tax wedge on total compensation, it simultaneously encourages overconsumption of vehicles, with implications for transport-related externalities and welfare.
By Carl Berry; VTI, LiU    Maria Börjesson; VTI, LIU    Jos van Ommeren; Vrije Universiteit Amsterdam
Presented by: Jos van Ommeren, Vrije Universiteit Amsterdam
Session 5: Urban Public Transport: Appraisal and Design June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 13
Session Chair: Paolo Beria, Politecnico di Milano
The Spatial Economics of Urban Transit in Santiago: Efficiency vs. Equity Abstract Urban sustainability relies heavily on public transport, which is frequently supported by operational subsidies, congestion pricing, and dedicated infrastructure like Bus Rapid Transit (BRT). While existing literature provides robust frameworks for analyzing these tools, it often assumes a representative, uniform city network. This study expands upon the Basso and Silva (2014) framework by introducing a novel spatial dimension to evaluate the relative efficiency and distributional impacts of transport policies in a segregated Latin American context, calibrated specifically for Santiago, Chile. By modeling geographical heterogeneity and transit corridors feeding a city center, we simultaneously assess allocative efficiency (total welfare) and equity (consumer surplus). Results in a centralized network show that while BRT maximizes welfare, it induces severe private vehicle congestion. Congestion pricing, alternatively, significantly improves system-wide travel speeds. When expanding to the spatial model, the optimal policy mix reveals a stark efficiency-equity trade-off. Implementing congestion pricing across both corridors and the center maximizes total welfare. However, this theoretical optimum requires eliminating subsidies, which drastically reduces consumer surplus and raises acute equity concerns. Conversely, maintaining or increasing subsidies protects consumer surplus but diminishes overall allocative efficiency. Furthermore, building BRT infrastructure in corridors consistently increases welfare regardless of the subsidy level, though marginal benefits soften where high-speed Metro alternatives exist. Ultimately, if distributional realities render subsidy reductions politically infeasible, we conclude that combining BRT expansion with congestion pricing offers the most pragmatic approach to optimizing urban transport performance without sacrificing equity.
By Leonardo J. Basso; Universidad de Chile    Hugo Silva; Pontificia Universidad Católica de Chile
Presented by: Hugo Silva, Pontificia Universidad Católica de Chile
Can pharmaceutical transport support the transformation to flexible transport systems? Abstract This study examines whether integrating pharmaceutical freight (medication deliveries from pharmacies) into passenger-oriented demand-responsive transport (DRT) systems can enhance viability in urban-rural transition zones, amid demographic aging and pharmacy access challenges. Employing MATSim simulations in the Göttingen region (urban core plus adjacent rural municipalities), the authors calibrate freight demand from a pharmacy survey revealing urban concentration with extensions into rural districts. Passenger demand follows established DRT patterns, using vehicles capped at 8 passengers/packages and a strict 20-minute waiting cap (exceeding requests rejected). Unregulated integration results reveal an "Urban Density Trap": as pharmaceutical delivery demand grows, the routing algorithm prioritizes short, high-frequency urban requests due to core pharmacy density. This crowds out rural requests (longer distances, lower pooling potential), causing sharply rising passenger waiting times, declining passenger rides, and disproportionately high rural rejection rates. While pooling improves and empty distance drops (operator efficiency gains), passenger metrics deteriorate significantly compared to pure passenger DRT, and per-pharmacy delivery quality dilutes. Fleet scaling restores delivery performance and enables full request service with marginal system changes versus baseline, yielding dramatic passenger improvements (lower waits, more rides)—aligning with prior integrated on-demand research. Preliminary geofencing (restricting urban-core pickups) indicates potential to redirect capacity ruralward and alleviate the trap, though full evaluation is ongoing. Overall, small-scale unregulated integration fails to ensure adequate supply in rural-transitional areas, risking inequities for elderly users reliant on medical mobility, despite utilization benefits. Targeted regulations or scaling are essential.
By Andreas Bossert; HAWK Holzminden    Alexander Röhrs; Hochschule für Angewandte Wissenschaft und Kunst    Jan Schlüter; HAWK Holzminden
Presented by: Alexander Röhrs, Hochschule für Angewandte Wissenschaft und Kunst
Demand, cost and benefits of urban mass transit. A review of a decade of feasibility studies in Milan Abstract The paper aims at providing a comparative and critical review of a more-than-a-decade long experience of Milan city in feasibility studies for large urban public transport projects. All studies insist on a well-defined territory – the metropolitan area of Milan, Italy – and have been conducted in a very homogeneous way from 2016 until today. The panel, consisting of 24 distinct feasibility studies for a total of approximately 190 alternatives, includes both selected and discarded alternatives. Such a large number and variety of studies, all including economic assessment through cost-benefit analysis, allow us to discuss the reasons behind socially worthwhile projects, such as the demand thresholds that justify costs, the relative importance of environmental externalities reduction, and the role of generated demand vs. existing one. A secondary aim of the study is then to provide reference values for costs, demand and parameters for European public transport studies. The paper, thanks to empirical evidence, ultimately provides policy indications for planners in mature cities, as a sort of guideline to design “good projects” in function of the sources of benefits realistically expected.
By Paolo Beria; Politecnico di Milano    Tommaso Battilocchi; Politecnico di Milano
Presented by: Paolo Beria, Politecnico di Milano
Session 6: Slots and Capacity Allocation June 17, 2026 13:30 to 15:00 Location: Palazzo Baroni 14
Session Chair: Achim Czerny, Hong Kong Polytechnic University
Do Incumbent Carriers Underuse Scarce Airport Slots? Evidence from Newark Airport’s Level 2 Announcement Abstract This paper examines whether incumbent carriers engage in slot hoarding at slot-constrained airports by strategically leaving operating opportunities unused in order to limit rivals’ access. It exploits the Federal Aviation Administration’s announcement on April 1, 2016 that Newark Liberty International Airport (EWR) would be redesignated from Level 3 slot control to Level 2 schedule facilitation, effective October 30, 2016. Because the FAA indicated that under Level 2 coordination it would place substantial weight on carriers’ prior approved schedules and actual operations (“track record”), the announcement plausibly altered incumbents’ incentives during the transition period. Using U.S. DOT On-Time Performance data, we construct a daily panel at the airport × 30-minute time-block × carrier level and estimate a difference-in-difference-in-differences (DDD) model. The design compares dominant and non-dominant carriers within time blocks at EWR before and after the announcement, relative to the corresponding dominant–non-dominant difference at JFK, LGA, and DCA, which did not experience a comparable redesignation announcement during the same period. The analysis distinguishes increased operations from increased competition by examining four outcomes: flights operated, within-block carrier shares, average seats per movement, and the number of effective competitors. This structure allows us to assess whether incumbents merely intensify use of existing capacity or whether rivals also gain greater access. By leveraging a clearly dated regulatory announcement, the study provides a more direct, incentive-based test of slot hoarding than proxy-based approaches and clarifies the competitive implications of reforms in slot-constrained airport settings.
By Hideki Fukui; Ehime University    Jules Yimga; Embry-Riddle Aeronautical University
Presented by: Hideki Fukui, Ehime University
Auctioning Congested Service Networks with Application to Air Traffic Control Markets Abstract Many industries requiring large and extensive networks are served by monopolistic infrastructure providers. Such networks may often suffer from congestion during peak periods. We analyse the effects of auctioning congestable service provision for the market in the case of natural monopolies, where competition in the market is not an option. The procurement auction includes a reward and penalty scheme as a function of service quality. We characterize the properties of the auction mechanism in which oligopolistic customers choose optimal combinations of servers, creating a network congestion game where serial and parallel competition play a role. The model is subsequently applied to the Western European air traffic control market, which handles 50\% of the European airline traffic. EU Member States each introduce a procurement auction and air navigation service providers bid to operate air traffic control in one or more regions for a pre-specified period of time. The results suggest that introducing competition for the market via outsourcing service provision leads to social cost efficiency and a substantial reduction in charges. In addition, the incentives are likely to lead to defragmentation of the European air traffic control market, were providers to win auctions in adjacent regions. We note that it is important to ensure at least two competitors participate for the auction process to remain successful over time hence suggest a cap on the market share served by any one geographical monopolist.
By Nicole Adler; Hebrew University of Jerusalem    Eran Hanany; Tel Aviv University    Stef Proost; KULeuven
Presented by: Nicole Adler, Hebrew University of Jerusalem
Invest in centralized capacity and get rid of traffic distribution rules? Abstract This study examines whether investments in centralized capacity can be a substitute for traffic distribution rules in congested multi-facility transport systems. Building on Czerny and Fukui (2024), which distinguish between quantity effects (excessive overall utilization) and misallocation effects (excessive utilization of specific facilities), the paper analyzes traffic distribution rules as a second-best policy instrument that improves welfare by correcting misallocation without reducing total traffic levels. The model considers two horizontally differentiated, asymmetric decentralized facilities (e.g., airports) sharing a congested centralized facility (e.g., airspace). Asymmetries arise in both capacity and demand: the facility with lower capacity is more desirable and therefore more congested under laissez faire. The framework captures key features of multi-airport regions such as New York, where congested airspace and perimeter rules at LaGuardia Airport coexist. The theoretical analysis combines general functional forms, closed-form solutions, and numerical simulations. The central question is whether expanding centralized capacity (e.g., air traffic management) can replace traffic distribution rules, thereby allowing policymakers to abandon politically contentious regulations such as perimeter restrictions. Preliminary numerical results suggest that, although traffic distribution rules do not achieve first-best utilization, they can substantially enhance welfare by correcting misallocation. In calibrated simulations reflecting realistic parameter values, such rules can approximately double welfare relative to laissez faire. The findings highlight the interaction between centralized and decentralized congestion and clarify the conditions under which infrastructure investment may—or may not—serve as a substitute for regulatory traffic distribution policies.
By Achim Czerny; Hong Kong Polytechnic University
Presented by: Achim Czerny, Hong Kong Polytechnic University
Session 7: Parking Markets and Curbside Regulation June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 8
Session Chair: Stefan Tscharaktschiew, TU Dresden
Charging is the new parking: The optimal density of on-street charging stations in a residential area Abstract Electric vehicles (EVs) are widely regarded as an important pathway to decarbonization, yet their adoption remains low. One of the barriers is insufficient public charging infrastructure. Early adopters have mostly relied on home charging. In contrast, mainstream adopters lack access to home charging, making public curbside charging essential to promote adoption. Converting curbside parking spaces into charging stations reduces parking capacity. If charging stations are vacant, cruising for parking time cost increase, especially for conventional vehicle users because they users cannot park in these spaces. This study examines the optimal density of curbside charging stations in urban residential areas, considering the tradeoff between the benefits and costs of converting conventional parking spaces into charging spots. A dynamic economic model is developed in which residents choose between being permit and non-permit holders for parking, and whether to park or to charge, based on private benefits and expected generalized cruising costs. Visitors arriving at a similar time also face a similar cruising time cost. Charging stations are restricted to EV users. The model is calibrated using German data. Preliminary findings suggest that the higher the charging density, the fewer permit holders there are. The expansion of charging infrastructure initially improves welfare by enabling EV charging and reducing cruising time costs. At the optimal level, cruising for parking is minimized, and households gain the most significant benefit from parking and charging. Expanding the charging stations beyond that point reduces welfare due to increased parking searches and a loss of consumer surplus of visitors.
By Nunnapas Saereeporncharenkul; TUD Dresden University of Technology    Stefan Tscharaktschiew; TU Dresden    Georg Hirte; Technische Universität Dresden
Presented by: Nunnapas Saereeporncharenkul, TUD Dresden University of Technology
Spatial Price Competition of Parking Garages and Curbside Regulation Abstract This paper investigates spatial price competition between off-street garages and their interaction with curbside regulation. We develop a theoretical framework using a taxicab distance metric to account for orthogonal urban grids, product differentiation, and the endogenous choice of cruising. Our model identifies a Parking Hub Effect, where localized clustering increases aggregate reliability. Drivers follow a two-stage process: first selecting a high-density Hub to minimize search uncertainty, then choosing a facility. This creates a Paradox of Price-Increasing Competition, where supply expansion can increase equilibrium prices by pulling drivers from the curb. Empirically, we test these findings in Lisbon using a spatial RDD and an IV approach. We exploit the historical path-dependency of former religious estates (convents) as a robust instrument for supply. In Lisbon’s dense medieval fabric, these unified estates provided the rare land-assembly advantage necessary for large-scale underground construction (68% of private garages), which remains physically impossible on fragmented private parcels. Our results resolve theoretical ambiguities: while the entry of independent rivals has a statistically insignificant negative effect due to low differentiation, we find robust evidence of market power through network concentration. Increasing facilities within the same internal network leads to significant price increases. Furthermore, garage prices are strategic complements to curbside fees and curbside capacity increases decrease garage prices. Our RDD confirms a significant price jump at regulatory boundaries, proving that municipal curb management is the primary determinant of off-street pricing and a vital tool for urban sustainability.
By Sofia F. Franco; University of California Irvine    Albert Gragera; Universitat Autònoma de Barcelona
Presented by: Sofia F. Franco, University of California Irvine
The Economics of Residential Parking Abstract In many cities around the world, the residential parking market is managed inefficiently. City officials grant residents parking permits that allow them to park their cars on public roads (on-street) in the vicinity of their homes virtually for free. The preferential treatment of residential parking causes various inefficiencies: it exacerbates cruising for parking, discourages visits of local shoppers and others, and, in the long term, it promotes private car ownership. This results in excessive parking search times, road congestion, pollution, loss in consumer surplus etc. In this paper we develop the first micro-based economic equilibrium model of residential parking. The model captures the dynamics of the residential parking market by considering the cumulative nature of arrivals of residents throughout the day. We consider a spatially homogeneous area which serves as parking location for (local) residents and (non-local) occasional visitors. Residents and visitors can choose between two parking locations, on-street parking and off-street parking (garage). Residents and visitors compete for parking and impose negative parking search externalities depending on time of arrival. The parking location choice of residents is based on a random utility approach. We calibrate the model to German data. We find that welfare losses from subsidized parking permits are much higher than previously suggested. The efficiency loss is about € 10 per permit and day. The marginal external cruising cost in the residential parking market is significant relative to the current price of a residential parking permit (around factor 50).
By Stefan Tscharaktschiew; TU Dresden    Nunnapas Saereeporncharenkul; TUD Dresden University of Technology    Georg Hirte; Technische Universität Dresden
Presented by: Stefan Tscharaktschiew, TU Dresden
Session 8: Transit Theory and Design Principles June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 9
Session Chair: Sergio Jara-Diaz, Universidad de Chile
Functional form assumption without apology: revisiting Parry and Small’s model Abstract This paper examines the sensitivity of optimal transit fare and subsidy calculations to functional form assumptions in the influential Parry and Small (2009) model. While subsequent applications of this model have invariably assumed isoelastic (log-linear) demand functions to estimate optimal subsidies, the robustness of this choice has never been formally assessed. Following the merger simulation literature, we specify three alternative demand systems — linear, logistic, and the Almost Ideal Demand System —– and recalculate optimal transit subsidies for the same three cities (Washington DC, Los Angeles, and London), modes (rail and bus), and time periods (peak and off-peak) as in the original study. Surprisingly, optimal subsidy rates are virtually identical across all demand specifications in 11 of 12 simulations. The sole exception arises when the optimal fare is significantly above the current fare, a case driven by a built-in model discontinuity related to passengers switching to timetable-based behavior when headways exceed 15 minutes. We show this robustness stems from: (i) the transit fare representing only a small fraction of the generalized price of travel; and (ii) compensating effects between marginal operating costs and marginal scale economies that dampen demand differences across specifications. Our results suggest that researchers applying the Parry and Small framework can confidently estimate optimal fares and subsidies without undue concern about the choice of demand specification, except when optimal fares substantially exceed current levels.
By Andres Gomez Lobo; University of Chile
Presented by: Andres Gomez Lobo, University of Chile
Last Train Departure Choice Problem with Agglomeration Economies Abstract This paper deals with the choice of the last train departure time by transit operator and by social planner. We develop a model of endogenous scheduling in which commuters choose the time to depart CBD to go home, taking into account the trade-off between benefits from activities in the CBD and leisure time at home. The benefit from activities in the CBD depend on how many people stay there at the same time. We numerically investigate the effect of parameters on the choice of last train departure. We also conduct empirical analysis utilizing the data from railway lines in Osaka metropolitan area and shows that variations of last train departure times are consistent with predictions of our model.
By Se-il Mun; Doshisha University
Presented by: Se-il Mun, Doshisha University
Effects of omitting users’ costs on public transport design: lessons from stylized networks. Abstract The primary trade-off faced in public transport planning and design is between the level of service experienced by users and the costs borne by the operators. It has been shown that these latter weigh more when imposing self-financing policies at the expense of the relevance of users’ costs (their time), inducing smaller than optimal fleets of larger than optimal vehicles. This study analyzes the effects of ignoring user costs (UC) in the strategic design of public transport systems from a structural perspective, with a particular focus on spatial design. The problem is addressed using three simplified networks (cross-shaped, triangular, and Y-shaped), over which traditional line structures are compared: Feeder-Trunk, Hub-and-Spoke, Direct (avoiding transfers), and Exclusive (avoiding transfers and intermediate stops). For each network and line structure, operational variables —frequencies and bus sizes— are optimized under two objectives: minimizing operator cost (OC) and minimizing OC+UC. Four elements stand out in the analysis: transfers, idle capacity, excess distance traveled, and stopped utilization. The main result is that the optimal structure varies with the level and distribution of demand when OC+UC are minimized, but when UC is omitted, the structure that generally dominates is the Exclusive one, always operating with an undersized fleet of oversized buses running at lower frequencies than those of the OC+UC optimum. Differences with what is reported in the literature are discussed, motivating further analysis in more complex networks.
By Sergio Jara-Diaz; Universidad de Chile    Diego de Amesti; Universidad de Chile    Antonio Gschwender; Universidad de Chile
Presented by: Sergio Jara-Diaz, Universidad de Chile
Session 9: Road Pricing: Design, Elasticity, and Equity June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 11
Session Chair: Maria Börjesson, VTI, LIU
Designing acceptable city tolls: The role of revenue recycling mechanisms Abstract Urban transport generates substantial external costs from congestion, air pollution, noise, and accidents, creating significant economic burdens. Although tolls are considered an efficient market-based instrument to internalize these externalities, implementation remains rare. The lack of public support is viewed as one of the main reasons. This study examines how revenue recycling mechanisms influence support for a city toll, where four different revenue recycling mechanisms derived from a hypothetical referendum questionnaire are analyzed: (1) investment in sustainable mobility, (2) discretionary allocation by city policymakers, (3) equal redistribution, and (4) personalized public transport vouchers (cashback). The referendum voting will be analyzed in an ordered logit model, including explanatory variables of socio-economic characteristics, spatial, work-related, mobility patterns and attitudes, as well as preferences for mode and departure time choices. Preliminary results reveal substantial variation across revenue recycling mechanisms: support is highest when revenues are earmarked for sustainable transport and lowest when allocation is left to policymakers. Around half of respondents change their vote depending on the recycling mechanism. The ordered logit model shows that different socio-economic, mobility-related, work-related, attitudinal, and preference variables significantly influence acceptability of city tolls.
By Sonja Fessler; Vienna University of Economics and Business    Stefanie Peer; Vienna University of Economics & Business
Presented by: Sonja Fessler, Vienna University of Economics and Business
Review and Meta-Analysis of International Evidence on Toll Price Elasticities Abstract This paper reports a review and meta-analysis of international evidence on toll charge elasticities, largely drawn from peer-reviewed published studies. It also provides an authoritative review of fuel price elasticity evidence against which the toll elasticity evidence can be compared and of previous toll elasticity reviews. The meta-model, which explains variations in toll elasticities across studies, is based on one of the largest samples of toll elasticity evidence reviewed. It determines a number of influences on toll elasticities, with a novel contribution being the impact of route competition which is to be expected but not previously quantified. Other credible effects were estimated to income, population density, electronic toll collection, the introduction of tolls, the estimation method and whether elasticities were short or long run, and whether the elasticity related to cars or trucks. The absolute toll elasticities implied by the meta-model were plausible. They suggest that the conventional wisdom of short run toll elasticities in the range -0.2 to -0.3 is, on average, too large, but that the premium for long run effects implies toll elasticities at the top end and beyond the ‘typical range’ of 0 to -0.5. The study’s findings strongly suggest that toll elasticities are generally larger than fuel price elasticities which is to be expected.
By Richard Batley; University of Leeds
Presented by: Richard Batley, University of Leeds
Equity and Heterogeneity in Congestion Charge Payments Abstract We analyse the distribution of payments with respect to income, home and work location, commuting across the cordon, and other socio-economic characteristics in the two full-scale congestion charging systems in Stockholm and Gothenburg. We use unique population-wide Swedish micro-registry data linking all residents to their vehicles and to annual congestion charges paid during 2018–2020. The payments are weakly regressive, with Suits indices of −0.03 (Gothenburg) and −0.06 (Stockholm). However, the income gradient is several times smaller once we control for time-invariant individual-specific factors. There is substantial heterogeneity in annual payments within income groups—so large that the regressive/progressive classification becomes of limited practical relevance. Few individuals pay the charge on a regular basis, implying that equity assessments based on self-reported behaviour in a “normal week” or on single-day travel observations can be misleading. When controlling for individual-specific effects, commuting across the cordon is only weakly associated with annual payments.
By Maria Börjesson; VTI, LIU    Nanna Fukushima; University of Stavanger Business School
Presented by: Maria Börjesson, VTI, LIU
Session 10: Rail Demand and Modal Choice June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 12
Session Chair: Javier Asensio, Universitat Autonoma de Barcelona
Analysing passengers' preferences for high-speed rail: low-cost operators and modal substitution effects Abstract European high-speed rail (HSR) markets are undergoing a profound transformation driven by intramodal competition and the emergence of low-cost business models. While the liberalisation of the aviation industry is well-established, the effective opening of HSR markets is more recent, introducing structural challenges such as capacity saturation and threats to the financial sustainability of incumbent operators. This study investigates passengers' preferences for HSR by specifically examining the role of low-cost operators and potential modal substitution effects.The research employs a discrete choice model using data collected through a web-based, Pivot-type stated preference questionnaire focused on the Italian case study. This approach captures passenger heterogeneity by pivoting choice cards based on respondents' actual recent long-haul trips. The analysis utilizes a Latent Class Nested Logit (LCNL) framework to distinguish between transport modes and specific service alternatives, while segmenting the market into business and leisure classes.Expected results suggest that willingness to pay (WTP) for ancillary services—such as Wi-Fi, luggage flexibility, and silent areas—varies significantly by travel purpose and distance. The study aims to estimate the WTP for these unbundled attributes to evaluate the competitive implications of low-cost rail entrants. Furthermore, it will be relevant gain hints on whether the entry of low-cost HSR operators primarily triggers a modal shift from long-distance coaches or results in the cannibalization of existing rail market segments
By Nicolò Avogadro; University of Bergamo    Mattia Borsati; University of Barcelona    Mirko Giagnorio; Sapienza University of Rome    Martina Gregori; Universitas Mercatorum
Presented by: Martina Gregori, Universitas Mercatorum
The Impact of Train Delays on Mode Choice: Evidence from Germany Abstract Railway services are central to passenger transport and climate change mitigation, yet concerns about declining reliability have intensified. This raises the question of whether worsening punctuality undermines the role of railway services in passenger transport. In this paper, we use the example of Germany to examine whether train delays affect mode choice. We combine high-frequency mobility data originating from anonymized mobile phone signals with detailed scraped train delay data mapped to a shared network structure. To address the endogeneity of delays, we exploit exogenous variation in regional train delays induced by initial delays of long-distance trains at distant departure stations. Using this variation, we estimate the effect of regional train delay on the share of trips by rail at the district-of-departure level, focusing on medium-distance travel of 30 to 150 km. This approach provides novel evidence on the implications of railway punctuality for mode choice, with implications for transport policy.
By Dorothea Kistinger; Potsdam Institute for Climate Impact Research    Nicolas Koch; Potsdam Institute for Climate Impact Res
Presented by: Dorothea Kistinger, Potsdam Institute for Climate Impact Research
Travel demand impacts of Spain’s HSR network Abstract We quantify the effects on travel demand due to the existence of Spain’s High Speed Rail (HSR) network. In doing so, we take into account how such effects vary by income level. Travel flows are obtained from cellphone location data where each inter-provincial flow has been allocated to a specific transport mode. We estimate a three-stage transport model for flows at the provincial level. In the first stage, travel demand generation of trips outside the province is estimated. Then, a destination choice model allocates such flows among all other provinces. The third stage consists in the modal distribution, where four transport modes are considered (rail, car, air and bus/coach). The income impacts on the estimated travel flows are considered by means of a recalibration exercise. Once the model has been estimated, a simulation exercise is carried out consisting in bringing railway travel times back to the situation prior to the construction of the HSR network. By comparing travel flows by mode and income level, the effects of such network can be quantified. The results show that the existence of the HSR has almost doubled overall rail patronage for interprovincial travel. Given that the losses of other modes (mostly private car) amount to less than half those gains, 53% of rail travel is due to new demand generation. These results are decomposed by income quintiles.
By Javier Asensio; Universitat Autonoma de Barcelona    Anna Matas; Universitat Autonoma de Barcelona
Presented by: Javier Asensio, Universitat Autonoma de Barcelona
Session 11: Aviation Policy & Competition June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 13
Session Chair: Davide Scotti, Università di Bergamo
Who benefits from Google’s SERP? The impact of the DMA on the Air Travel Market Abstract This paper examines the impact of Google’s Search Engine Results Page (SERP) redesign implemented on March 6, 2024, in response to the Digital Markets Act (DMA), on web traffic to airline and flight comparison websites. The reform removed the Google Flights module and introduced new sections — “Airlines Options” and “Flight Sites”—designed to limit self-preferencing and to improve the visibility of third-party services. Using a Difference-in-Differences (DiD) approach and granular domain-level traffic data for four EU countries, with the United Kingdom and the United States as control markets, we estimate the causal effects of the SERP redesign over the period November 2023 to December 2024. The results show significant increases in organic search traffic to airline websites, driven mainly by low-cost carriers, smaller national airlines, and airlines with limited pre-reform visibility. Organic search traffic to low-cost carriers increased by 14-28% on desktop devices and by 8-22% on mobile devices, while network carriers and dominant national airlines experienced little or no effects. For flight comparison websites, traffic gains were concentrated among smaller platforms, with increases of up to 28.9%, whereas large platforms such as Skyscanner and Booking.com show no significant changes in organic traffic. Overall, the DMA-mandated SERP redesign generated strong redistributive effects, reallocating user attention towards smaller market participants and increasing contestability in digital air travel market.
By Xavier Fageda; University of Barcelona
Presented by: Xavier Fageda, University of Barcelona
Competing on Emissions Charges Abstract We develop a game-theoretic framework in order to estimate market equilibria in the presence of environmental policies at both regional and global levels. In a two-stage game, multiple regulators maximize social welfare within their jurisdiction by setting emission charges, while airlines compete through fleet choice, frequencies and fares given the choices of the first-stage regulators. Airline decisions include determining the extent to which environmental charges are absorbed or passed on to consumers and how fleets are upgraded and utilized across the network. The equilibrium outcomes reveal the presence of multiple potential distortions in aviation markets that could undermine the effectiveness of environmental policies. Applying the model to North American, Western European and Trans-Atlantic markets, we find that the most efficient mechanism involves a single regulator employing tax discrimination between regions. However, the analysis shows that welfare improvements may come at the expense of environmental goals because self-interested regulatory interactions limit policy effectiveness.
By Nicole Adler; Hebrew University of Jerusalem    Gianmarco Andreana; University of Bergamo    Gerben de Jong; VU Amsterdam
Presented by: Gianmarco Andreana, University of Bergamo
The Effects of Unlocking Intra-African Air Connections Abstract Aviation is crucial in Africa, a vast continent where alternative modes of transportation are often limited. Yet intra-African air travel remains underdeveloped. Against this backdrop, the continent has pursued several liberalization initiatives, most prominently the Yamoussoukro Decision (1999) and, more recently, the Single African Air Transport Market. However, implementation remains incomplete and international markets are still largely shaped by bilateral air service agreements. In this context, we study a specific and policy-relevant dimension of liberalization: the opening and use of fifth freedom traffic rights. We use a monthly panel of intra-African international nonstop passenger services operated by the top 20 African carriers over 2015–2024, excluding the COVID period (2020–2022). Adopting a multi-leg perspective, we identify fifth-freedom operations by tracking itineraries O−G−D in which a carrier based in country O sells and operates the onward leg G−D with destination D in a third country, and assign treatment to the corresponding feeder segment O−G. We then implement a segment-level difference-in-differences design comparing treated feeder segments to segments that never become associated with a fifth-freedom operation. Consistent with a network-capacity mechanism, our results indicate that fifth-freedom openings reduce fares on treated feeder segments while increasing feeder capacity, with passenger volumes broadly unchanged and load factors declining. We also find heterogeneity by onward-leg distance and gateway type, and expect to find evidence of gateway spillovers: following a fifth-freedom opening on G−D, flight frequencies into gateway G is expected to increase from other routes, consistent with G becoming a more attractive connection point.
By Jan K. Brueckner; University of California, Irvine    Andrea Gualini; Università di Bergamo    Laura Ogliari    Davide Scotti; Università di Bergamo
Presented by: Laura Ogliari,
Session 12: Airline Schedule Competition and Cooperation June 17, 2026 15:30 to 17:00 Location: Palazzo Baroni 14
Session Chair: Gianmaria Martini, Università di Bergamo
Strategic Scheduling and Mutual Forbearance: The Moderating Role of Intermodal Substitution in European Aviation Abstract Multimarket contact (MMC) facilitates competition softening through mutual forbearance, a phenomenon well-documented in US and Chinese aviation but largely unexplored regarding service attributes in the European market. This study addresses this gap by investigating the extent to which MMC facilitates forbearance in flight scheduling decisions and how this relationship is moderated by intermodal competition from High-Speed Rail (HSR). The research focuses on the unique structural features of European aviation, including node-centric networks, airport slot constraints, and extensive rail substitutability. We posit that while high-MMC environments facilitate temporal differentiation in daily schedules and synchronous seasonal adjustments, these effects are conditioned by the Generalised Journey Travel Time (GJTT) ratio between air and rail. Using a fixed-effects panel regression model and data sourced from OAG and GTFS (2022–2025), we evaluate two novel metrics: Daily Scheduling Differentiation and Seasonal Profile Dissimilarity. Anticipated results suggest that MMC promotes mutual forbearance through increased schedule intervals. However, this coordination is significantly attenuated as the Air/Rail GJTT ratio approaches parity, indicating that intermodal substitutability undermines the retaliatory mechanisms required for tacit coordination. This study provides a novel global contribution by being the first to examine strategic scheduling forbearance. The findings offer a pro-competitive rationale for HSR infrastructure investment and provide an empirical basis for antitrust authorities to incorporate rail connectivity into merger assessments, ultimately enhancing the calculation of consumer welfare in transport network evaluations.
By Lewis Shanahan; Delft University of Technology    Mahnam Saeednia; Delft University of Technology    Patrick Stokkink; Delft University of Technology    Oded Cats; Delft University of Technology
Presented by: Lewis Shanahan, Delft University of Technology
Measuring airline competition through schedule differentiation Abstract We present a novel method for measuring airline competition by analyzing the temporal proximity of departure times across carriers. Examining 245,214 domestic flights, we find that price competition is strongest when competing flights depart within an hour of each other. A competing flight scheduled within 30 minutes lowers fares by approximately $30, while a 1-hour window reduces fares by about $19. Temporal competition from low-cost carriers has a greater impact on fares than full-service carriers. Our findings highlight that regulators should incorporate measures of schedule-based competition when evaluating mergers in industries where timing plays a pivotal role. THE FULL PAPER HAS BEEN UPLOADED
By Marco Alderighi; Univeristy of Milan    Alberto Gaggero; University of Aosta Valley    Alexander Luttmann; The MITRE Corporation
Presented by: Alberto Gaggero, University of Aosta Valley
Effects of Airline Cooperation with local and global markets Abstract This paper develops a theoretical model of airline cooperation between two carriers operating complementary flight segments. Passengers are divided into local travelers, who buy single-leg flights, and connecting travelers, who purchase the full itinerary. Each airline chooses price and service quality under three regimes: non-cooperative competition, cooperation without price discrimination, and cooperation with price discrimination. Under competition, double marginalization leads to excessive prices and insufficient quality. Cooperation internalizes cross-segment externalities, yielding higher quality and lower prices, particularly when quality costs are high. Allowing price discrimination across market segments further increases joint profits and can reduce connecting fares. The analysis provides analytical equilibrium solutions and comparative statics on cost and demand parameters. It shows that price discrimination should be carefully monitored by antitrust authorities since there are many market situations where social welfare is higher if charging different prices to local and global passengers is forbidden.
[slides]
By Gianmaria Martini; Università di Bergamo
Presented by: Gianmaria Martini, Università di Bergamo
Session 13: Air–HSR Competition under Liberalization June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 8
Session Chair: Albert Gragera, Universitat Autònoma de Barcelona
Air–HSR competition in the era of high-speed rail liberalization: intermodality, pricing dynamics, and sustainability Abstract This paper investigates how airlines adjust prices and capacity in response to both intra-modal airline competition and inter-modal competition from high-speed rail (HSR), with particular attention to the recent liberalization of European HSR markets. While the effects of airline deregulation and low-cost carrier entry are well documented, much less is known about how the entry of additional rail operators reshapes competitive dynamics between air and rail transport. Understanding these responses is increasingly relevant as liberalization expands HSR supply and intensifies multimodal rivalry on short- and medium-haul routes. We assemble a novel dataset combining airline supply data from RDC Aviation with detailed information on HSR services across mainland EU routes between 2017 and 2025. The dataset provides route-level fares, flight frequencies, and the number of active airline and HSR operators. Competition is measured through operator counts and discrete indicators capturing different levels of market structure. We estimate two-way fixed-effects panel models and staggered difference-in-differences to isolate the causal impact of competitive pressure on airfares and capacity. Preliminary results show that full-service carriers respond strongly to HSR entry, reducing both fares and flight frequencies in a manner comparable to their reaction to low-cost airline competition. In contrast, low-cost carriers show limited or no systematic adjustments. Moreover, the competitive impact of HSR appears stronger on relatively longer routes, suggesting that rail liberalization may have expanded the effective distance range over which air–rail competete.
By Mattia Borsati; University of Barcelona    Xavier Fageda; Universitat de Barcelona    Hana Fitzová; Masaryk University    Vilem Paril; Masaryk University
Presented by: Mattia Borsati, University of Barcelona
Fewer seats, resilient frequencies: Impacts of large-scale High-Speed Rail liberalisation on air transport supply Abstract This paper examines the liberalisation of high-speed rail (HSR) as a driver of intermodal competition in long-distance passenger transport. While previous research has primarily focused on the effects of HSR infrastructure deployment on aviation, less is known about how opening HSR markets to new entrants reshapes this competition. Drawing on the Spanish HSR liberalisation in 2021, we are the first to evaluate the causal effects of a large-scale liberalisation on competing air transport supply. Using a regression discontinuity design, we find significant long-term reductions in airline seat supply (10-16%), but limited impact on frequencies. We then uncover two mechanisms underlying these results. First, airlines’ primary response was to down-gauge aircraft. Second, a market share emerged from the need of legacy carriers to preserve frequencies to feed their hub. Our results underscore the broad and significant implications of liberalisation for intermodal substitution
By Daniel Albalate; University of Barcelona    Albert Gragera; Universitat Autònoma de Barcelona    Pere Suau-Sánchez; Universitat Oberta de Catalunya
Presented by: Albert Gragera, Universitat Autònoma de Barcelona
Session 14: Autonomous Vehicles in Transit Systems June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 9
Session Chair: Manon BASSEVILLE-ALLOT, LVMT - Université Gustave Eiffel
Autonomous Transit Adoption Incentives, Policy and Investment: A Multi-stage Stackelberg Equilibrium Analysis Abstract Game theory provides a framework to analyze strategic interactions among stakeholders in the emerging autonomous mobility ecosystem. While prior studies have explored various game-theoretic approaches to examine strategic decisions regarding autonomous vehicle (AV) adoption, infrastructure investment, and policy development, the role of autonomous transit (AT) in mixed traffic remains underexplored. This study develops a three-stage Stackelberg game with endogenous passenger demand to analyze the strategic adoption of AT within mixed traffic environments. We examine how congestion externalities, emission pricing policies, and passenger preferences collectively influence AT adoption and overall system performance. There are four types of players involved in this game: (i) a transportation policymaker, who sets congestion pricing, emission regulations, and vehicle-to-infrastructure connectivity standards; (ii) a transit agency, who determines pricing strategies and service levels for both AT and conventional human-driven transit (CT) while minimizing its operational cost; (iii) AV and human-driven conventional vehicle (CV) producers, who set fixed vehicle prices; and (iv) passengers, who choose among AT, conventional transit, autonomous vehicles, and human-driven vehicles to maximize individual utility. Unlike existing AV adoption models, this framework explicitly endogenizes autonomous transit as a strategic mode and links passenger adoption to congestion pricing, emission policy, and V2I investment in a unified equilibrium.
By Yili Tang; University of Western Ontario    Jingwen Yao; Western University
Presented by: Jingwen Yao, Western University
Context-Dependent Valuation of Autonomous Shuttle Travel Time: A Stated Choice Experiment Comparing Monomodal and Intermodal Trip Contexts Abstract Autonomous shuttles are increasingly deployed for first-last mile connectivity, yet economic assessments assume time valuations transfer directly from isolated to integrated multimodal contexts—an assumption never empirically tested. We challenge this using a dual-experiment stated choice design examining how trip context shapes autonomous shuttle perception: travelers may value shuttles differently when they serve as access modes versus standalone alternatives. Drawing on 600 respondents across Bayesian D-efficient designs, we test whether favorable AV time valuations documented in isolated settings persist when shuttles function as access links within multimodal networks. Our findings directly address a critical contradiction: while monomodal studies show 30% VTTS reductions for autonomous driving, recent evidence paradoxically finds AVs as last-mile access modes valued worse than manual cars. By revealing how integration within multimodal chains fundamentally alters temporal perception, we provide essential valuation parameters for cost-benefit analyses and expose limits to transferring behavioral insights across trip configurations—critical for deploying autonomous shuttles in complex transport networks.
By Manon BASSEVILLE-ALLOT; LVMT - Université Gustave Eiffel    Vincent benezech; LVMT    Nicolas Coulombel; Ecole des Ponts ParisTech
Presented by: Manon BASSEVILLE-ALLOT, LVMT - Université Gustave Eiffel
Session 15: Distributional Effects and Tradable Permits June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 11
Session Chair: Anke Ye, Technical University of Munich
Distributional Effects of Distance-Based Road Pricing: A Behavioral Microsimulation Study for the Brussels-Capital Region Abstract Distance-based road pricing has long been recognized as an efficient instrument to internalize the external costs of car use. However, its implementation in passenger transport remains limited, largely due to concerns about its equity impacts. This paper provides an ex-ante evaluation of the distributional effects of a distance-based tax in the Brussels-Capital Region, explicitly accounting for both static and dynamic effects. Methodologically, the study adopts a behavioral microsimulation approach based on the construction of a synthetic microdataset that integrates three data sources: mobility survey data, administrative fiscal records, and estimated individual behavioral responses derived from a stated preference discrete choice experiment. Individual elasticities are estimated using a Mixed Multinomial Logit model. Preliminary findings suggest that, in aggregate, a distance-based tax is not necessarily more regressive than the current annual vehicle ownership tax, primarily due to lower vehicle ownership among low-income households. However, regressivity emerges when focusing on drivers, reflecting both uniform pricing structures and higher per-kilometer costs associated with less efficient vehicles used by lower-income groups. Initial computations further suggest stronger behavioral responses among these groups, raising important questions about policy designs that reconcile efficiency objectives with equity considerations. This research contributes to the literature on transport pricing, particularly road taxation, as a strategy to address vehicle-use externalities. To that end, by incorporating individual-level behavioral responses, it addresses key empirical gaps in the literature by capturing behavioral adaptation, a critical yet underexplored determinant of the impacts of distance-based taxation.
By Jean Paul Madrigal Rodriguez; Université catholique de Louvain
Presented by: Jean Paul Madrigal Rodriguez, Université catholique de Louvain
Decarbonizing Last-Mile Delivery: An Agent-Based Evaluation of Tradable Credit Schemes Abstract This study investigates the implementation of Tradable Credit Schemes (TCS) in urban logistics, comparing an integrated mobility-delivery market (MoLoCo) against a standalone logistics market (LoCo). Using an agent-based simulation of Munich, Germany, the research evaluates how these regulatory frameworks influence both market stability and provider behavior. The findings reveal a trade-off: the MoLoCo framework benefits from a "Coupling Effect" that stabilizes credit prices but inadvertently triggers a "Green Logistics Paradox" by encouraging online shopping. Conversely, the LoCo model accelerates decarbonization through high credit prices but risks a "Reverse Modal Shift," where increased delivery costs drive consumers back to physical stores, worsening traffic. While fleet electrification reduces costs, only cargo bikes effectively address urban congestion. The paper concludes by recommending a phased roadmap: starting with sector-specific regulation to force technological shifts, followed by a transition to an integrated market to optimize social welfare.
By Anke Ye; Technical University of Munich    Roman Engelhardt; Technical University of Munich    Klaus Bogenberger; Technical University of Munich
Presented by: Anke Ye, Technical University of Munich
Session 16: Commuting and Refueling Choices June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 12
Session Chair: Romain Gaté, Paris Dauphine University
Time, Price, Brand? Determinants of Individuals’ Gas Station Choice Abstract Car drivers frequently decide where to refuel in a market with many spatially differentiated yet otherwise nearly homogeneous alternatives. Although economic theory predicts largely rational choices based on distance and price, empirical evidence on individual gas station choice is scarce. This project adopts a consumer-centered perspective to analyze the determinants of gas station choice and their variation across socio-economic groups, contributing to debates on fuel market regulation and charging infrastructure expansion. The analysis relies on a dataset combining GPS tracking and survey data from about 5,000 participants in Germany (April-December 2023). Refueling stops are identified by matching movement data with the locations of roughly 17,000 gas stations. For 938 refueling decisions by 385 individuals, individualized choice sets are constructed based on deviations from the direct route between trip origin and destination, enriched with routing information and real-time fuel prices. A random utility-based discrete choice model assesses the effects of driving time and fuel price, including interactions with socio-economic characteristics and gas station attributes. Preliminary results highlight driving time as the most consistent and statistically significant determinant, indicating a strong preference for stations close to the direct route. Price effects are generally negative but less robust across model specifications, underscoring the need for further refinement. Ongoing work focuses on finalizing the model specification and conducting comprehensive robustness checks to derive meaningful and policy-relevant conclusions.
By Dennis Gaus; TUD Dresden University of Technology    Heike Link; DIW    Juan de Dios Ortuzar; PUC
Presented by: Dennis Gaus, TUD Dresden University of Technology
Urban form and peak-hour commuting time: case study with French urban areas. Abstract We estimate the effects of urban form on home-to-work commuting times (during peak hours) within French urban areas in 2014. A higher geographical concentration of jobs relative to population within urban areas appears to significantly influence commuting times. However, our estimates suggest relatively small effects. Average travel times between place of residence and place of employment would decrease by 10% if the spatial distribution of jobs were identical to that of the population within urban areas. Our results show that home-to-work commuting times depend on many factors other than the spatial distribution of jobs within urban areas (density, demographics, public transport, etc.).
By Romain Gaté; Paris Dauphine University
Presented by: Romain Gaté, Paris Dauphine University
Session 17: Advanced Air Mobility: Markets and Acceptance June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 13
Session Chair: Naomi Sieben, Netherlands Aerospace Centre (NLR)
Cost, Competition, and Demand: An Integrated Economic Model of the Advanced Air Mobility Market Abstract Advanced Air Mobility (AAM) has attracted growing attention as a potential complement to urban surface transport, yet substantial uncertainty remains regarding its economic viability, pricing, and market structure. Existing studies report widely dispersed adoption forecasts and often rely on exogenous assumptions about costs and accessibility. This paper develops an integrated economic framework to assess the conditions under which AAM services can achieve sustainable market adoption. We combine a detailed supply-side cost model for eVTOL operations with a discrete-choice demand model describing traveler mode choice between private cars, taxis, and AAM. Demand and supply are linked through a Bayesian calibration procedure based on observed travel patterns in Germany and benchmark operating costs. The model is calibrated to match literature-based mode shares, empirical cost elasticities, distance-related indicators, and a benchmark AAM operating cost, ensuring internal consistency between user behavior and operator costs. Results indicate that AAM adoption is strongly distance-dependent and concentrates on medium-to-long trips. The framework predicts equilibrium prices that are lower and demand levels that are higher than those reported in most existing studies, and these findings remain robust across a wide range of cost structures, elasticities, and competitive settings. Substantial fixed costs generate scale economies that favor concentrated market structures. An extension incorporating autonomous taxi services shows that emerging AV-based competition significantly affects AAM pricing and demand. Overall, the framework provides a transparent and empirically grounded tool for evaluating the economic viability of AAM in future urban mobility systems.
By Carlos Beltran; Dresden University of Technology    Georg Hirte; Technische Universität Dresden    Stefan Tscharaktschiew; TU Dresden    Erik Verhoef; VU Amsterdam
Presented by: Carlos Beltran, Dresden University of Technology
Psychological Drivers of passengers’ intention to use eVTOL eVTOL Air Taxis Abstract Electric Vertical Take-Off and Landing (eVTOL) aircrafts are increasingly discussed as a component of future Urban Air Mobility (UAM) ecosystems. While technical development is progressing rapidly, successful implementation depends not only on operational feasibility but also on user acceptance. Understanding psychological determinants of behavioral intention is therefore critical for system integration and deployment strategies. The present study investigates factors influencing passengers’ intention to use eVTOL aircraft using an extended Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) framework. The proposed model examines the role of performance expectancy, effort expectancy, hedonic motivation, price expectation, perceived unsafety, and social image in shaping behavioral intention to use eVTOL as a novel mode of transport. Data was collected through an online survey in which participants evaluated eVTOL. Measurement items were adapted from established acceptance research on emerging transport technologies and contextualized specifically for eVTOL operations. This study contributes to the systematic evaluation of human factors associated with emerging aviation technologies. The findings are intended to inform future research on passenger decision-making, guide communication and implementation strategies, and support evidence-based integration of eVTOL systems into multimodal transportation networks.
By Naomi Sieben; Netherlands Aerospace Centre (NLR)
Presented by: Naomi Sieben, Netherlands Aerospace Centre (NLR)
Session 18: Entry and Spatial Concentration in Airline Markets June 17, 2026 17:15 to 18:15 Location: Palazzo Baroni 14
Session Chair: Georg Hirte, Technische Universität Dresden
LOW-COST CARRIER ENTRY AND TRAFFIC EFFECTS IN ARGENTINA’S DOMESTIC AIR MARKET Abstract This paper analyzes the effects of the liberalization of Argentina’s domestic aviation market following the entry of low-cost carriers (LCCs) in 2018. Argentina represents a distinctive case of air transport liberalization, characterized by a mixed oligopoly in which a state-owned incumbent competes with private airlines, and by a partial liberalization process implemented selectively across routes. Using route-level supply data for the period 2017–2023, we examine LCC entry decisions and their impact on air service supply. First, we estimate a logit model to identify the route characteristics associated with LCC entry. Second, we causally assess the effects of LCC entry on flight frequencies and seat capacity using recent advances in staggered difference-in-differences methods. Our results show that LCCs are more likely to enter dense routes and that their entry leads to substantial increases in flight frequencies and seat capacity on routes that existed prior to liberalization. These findings suggest that the effects of LCC entry are primarily associated with traffic consolidation on existing routes rather than with the creation of new air services.
By Valeria Bernardo; Tecnocampus, Pompeu Fabra University    Germán Coloma; Universidad del Cema    Xavier Fageda; University of Barcelona
Presented by: Valeria Bernardo, Tecnocampus, Pompeu Fabra University
The effect of airport entry and exit on spatial market concentration in Europe Abstract This paper studies the causal link between airport entry and exit and market concentration. We apply a set of modern causal inference methods for staggered treatments. Our dataset includes European airports observed between 2000 and 2019. The Herfindahl-Hirschman Index (HHI) indicates that entry and exit occur in highly concentrated markets. We identify an increase in market concentration in seats and departures after entry and a decline after exit by applying seven quasi-experimental approaches that address staggered treatment. We test the robustness of our results by employing alternative definitions of catchment areas, considering international airports and all catchment overlaps between treated and event airports, assessing sensitivity with time and space placebos, and performing leave-one-out analysis. We further apply different catchment definitions, including or excluding international airports, consider exits and entries within existing catchment areas and those with overlapping catchments. Our findings robustly confirm that entry reduces market concentration, while exit increases it, for seats and departures in the treated catchment areas.
By Tolga Ülkü; IU International University of Applied Sciences    Wolfgang Grimme; DLR    Georg Hirte; Technische Universität Dresden    Sharavan Kumar; DLR    Hans-Martin Niemeier; University of Applied Sciences Bremen
Presented by: Georg Hirte, Technische Universität Dresden
Session 19: Open-Access Competition and Liberalization in European Rail June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 8
Session Chair: Pedro Cantos-Sanchez, Universitat de València
Regulatory Challenges in European High-Speed Rail Liberalisation Abstract The European Commission's Fourth Railway Package has catalysed liberalisation of domestic passenger rail markets from 2020, intensifying direct competition on national high-speed lines and cross-border connections. This empirical study scrutinises the regulatory frameworks governing operator entry and service integration across major European corridors. We analyse rules for market access, tendering standardisation for public service contracts, and infrastructure coordination, identifying a spectrum from open liberalisation to entrenched state monopolies. Key challenges emerge in enforcement: persistent infrastructure access discrimination by incumbents, anti-competitive agreements among operators, and abuses of dominant positions, as evidenced in cases adjudicated by national regulators (e.g., France's SNCF proceedings) and the European Commission. Additional hurdles include inconsistent vertical separation mandates, opaque track access charging, and fragmented cross-border authorisation processes, which undermine fair competition. Drawing on regulatory documents and 20+ competition authority investigations (2020–2025), the study maps these manifestations and proposes targeted reforms, such as harmonised EU access remedies and strengthened independent oversight, to address gaps in the evolving regulatory landscape for the European high-speed rail.
By Monika Jandová
Presented by: Monika Jandová,
THE IMPACT OF RAIL COMPETITION ON DEMAND Abstract Over the past three decades, the European Union has promoted the liberalization of passenger rail markets to enhance efficiency, service quality, and competitiveness. One key reform has been the introduction of open-access competition, allowing multiple operators to provide commercial services on the same routes. While competitive tendering has been widely studied, the demand effects of open-access competition remain insufficiently explored, particularly in cross-country settings. This paper investigates whether the introduction of open-access competition increases the modal share of passenger rail in European countries. Using panel data for 27 European countries over the period 2004–2023, the analysis applies a treatment–control framework combined with two-way fixed-effects models. Countries that introduced open-access competition—Austria, Italy, the Czech Republic (around 2012), and Sweden (later in the decade)—are compared with countries without significant open-access entry. The dependent variable is rail’s modal share in total passenger transport. The specification controls for GDP per capita, unemployment, rail fares, fuel prices, the share of high-speed rail infrastructure. Preliminary results indicate that open-access competition is associated with a positive increase in rail modal share, although the magnitude varies across countries and model specifications. The strongest effects are observed in Austria and the Czech Republic, while results for Italy are weaker. In contrast, the expansion of high-speed rail infrastructure shows a consistently strong and robust association with higher rail modal share. Overall, the findings suggest a positive but heterogeneous impact of open-access competition, alongside a more systematic role of infrastructure development in shaping passenger transport outcomes.
By Zdeněk Tomeš; Masaryk University
Presented by: Zdeněk Tomeš, Masaryk University
Competition and departure-time differentiation in High-Speed-Rail markets Abstract Open access competition has been introduced in many European rail markets. The effects in terms of prices and frequencies have been deeply investigated. However, there is hardly any research about the relevance of how departure time shapes the competition. Our goal is study how competition in departure times affects price levels in railway markets. We focus on the recently liberalized High-Speed-Rail market in Spain. There are four operators in which the incumbent Renfe offers two different services: Ave focused on higher quality services and Avlo, offering low-cost services. Additionally, Ouigo, a low-cost railway company belonging to the French operator SNCF, and Iryo (whith higher quality than Iryo). entered both markets. The paper is developed in two parts. Firstly, we define a Hotelling model where the spatial location corresponds to the departure time of a particular operator service. We characterize equilibrium prices, that depend on the identity of neighbor departures. Secondly, we test the predictions of the theoretical model using empirical data during 2022-2024. Data is obtained using a web crawler program that extracts price quotes from the operators’ websites. Preliminary results indicate that Ave prices in the economic fare fall when either Iryo or Ouigo are the companies with the closest departure time. Conversely, no significant effects are found when the train with the closest departure time is Avlo. If an Ave train departure is surrounded by Iryo, then the decrease in Ave prices is stronger. It therefore seems that interlaced departure times leads to more intense competition.
By Pedro Cantos-Sanchez; Universitat de València    Rafael Moner-Colonques; Universitat de València    Alba Ruiz-Buforn; Universitat de València    Jose Sempere-Monerris; University of Valencia
Presented by: Pedro Cantos-Sanchez, Universitat de València
Session 20: Free-Fare Transit Policies June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 9
Session Chair: Matheus Loureiro, University of São Paulo (USP)
Evaluation of Transportation Mode Choice with the Free-Fare Bus in Sao Caetano do Sul, Brazil Abstract Urban mobility is a fundamental component of the daily life in cities, considering it guarantees access to health services, employment, education and leisure. Free-Fare on Public Transport (FFPT), a policy that aims to make the public transport economically more accessible, has been implemented around the world to deal with the mobility issue. The outcomes of this policy remain a subject of ongoing scholarly debate, considering it impacts different areas, such as health, labor, and environment. Existing research in this field is largely situated within the context of understanding FFPT’s influence on travel mode selection. This research contributes to the debate on FFPT's effects on mode choice by focusing on a high-density environment. It investigates the city of São Caetano do Sul (SCS), in the Metropolitan Region of São Paulo, Brazil, which involves a small, but dense area. The FFPT in this city was implemented during the household survey “Origin-Destination Metro” data collection period, in 2023, and this is the main important database for the research. It also uses General Transit Feed Specification (GTFS) files to estimate the commuters’ paths by public transport. Using a discrete choice model, the research estimates the impact of this policy in the commuter’s mode choice. The study focuses on travels related to work and study and in the modes of car, public transport, motorcycle, and walk. Public transport includes an integrated network of subway, rail, different cities municipal bus and intermunicipal bus. Results indicate that the main important changes after the FFPT were a rise in public transport travels due to a decrease in motorcycle and car travels. Among different socioeconomic classes, low-income groups were the most benefited by this policy. Moreover, women and non-white people were proportionately more impacted than men and white ones. The findings suggest that FFPT can be an effective instrument not only for achieving modal shift but also for advancing urban equity in metropolitan areas.
By Aidan de Araujo; University of Sao Paulo
Presented by: Aidan de Araujo, University of Sao Paulo
Do Temporary Fare-Free Policies Create Lasting Ridership? Evidence from a Natural Experiment in Stavanger Abstract Cities increasingly experiment with temporary fare reductions—including fare-free public transport—to accelerate modal shift, yet it remains unclear whether short-run price interventions generate lasting changes in travel behavior. This paper examines whether a temporary fare removal produces persistent increases in public transport demand and whether demand responds asymmetrically to fare decreases versus subsequent fare reinstatement. We exploit a natural experiment in Stavanger, Norway, where residents were granted free access to public transport for six months in 2023. Using daily automatic passenger count data and a difference-in-differences design comparing Stavanger to surrounding municipalities, we estimate both short-run and post-intervention effects. We find a clear short-run demand response: total boardings in Stavanger increased by approximately 7.5–9.1% during the fare-free period relative to the control group. However, the increase does not persist after fares were reinstated; post-intervention estimates are slightly negative (around −2.6 to −3.3%) and not statistically distinguishable from zero across specifications. Overall, the results suggest that temporary fare reductions can boost ridership in the short run but are unlikely to induce durable habit formation without complementary measures such as service improvements or constraints on car use. The findings contribute to the literature on public transport demand, price elasticity, and the design of fare policies for sustainable urban mobility.
By Fredrik Alexander Gregersen; Insitute of Transport Economics
Presented by: Fredrik Alexander Gregersen, Insitute of Transport Economics
Next Station: Closing the Gap in Education? The Impact of Fare-Free Public Transport in São Paulo Abstract Although free public transport free-fare for students is a common policy, previous analyses in regions with low public transport use have shown inconclusive results. In 2015, São Paulo, Brazil, began offering free public transportation to public secondary school students, benefiting nearly 1.5 million teenagers. This study evaluates the impacts of this policy by examining short-term outcomes, including secondary education enrollment, completion, proficiency, and labor market participation, as well as medium-term outcomes, such as higher education enrollment, post-school labor market participation, and wages. Using a comprehensive dataset that tracks all Brazilian students and their formal labor market participation from 2007 to 2023, the analysis employs a difference-in-differences strategy. It compares São Paulo students, before and after the policy, with students from similar schools in other Brazilian cities. Heterogeneity is assessed by neighborhood-level public transportation accessibility. Preliminary findings show that the policy led to increases in both enrollment and dropout rates, with these effects more pronounced in schools with higher urban accessibility. These results indicate mixed short-term impacts, with policy effects varying based on school location. The proposed individual-level analysis has been approved and is scheduled to take place at the Brazilian Ministry of Education’s Protected Data Room between December 2025 and February 2026.
By Matheus Loureiro; University of São Paulo (USP)
Presented by: Matheus Loureiro, University of São Paulo (USP)
Session 21: Spatial Economics of Transport and Mobility June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 10
Session Chair: Guillaume Monchambert, University of Lyon
Highway to Heaven? The Effects of a Rural Transport Infrastructure Improvement on Local Accessibility and Economic Activity Abstract Transport infrastructure plays an integral role in regional development, particularly in rural areas, where access to socio-economic opportunities is often limited. Understanding the effects of rural transport policy on local populations is therefore central to addressing growing urban-rural divides and regional disparities concerning economic, demographic, social, and political outcomes in recent years. This thesis examines the economic impacts of the Waldviertel Highway, a proposed but unimplemented transport infrastructure project in the region of Waldviertel in rural Austria. Short-term effects on local socio-economic opportunities are evaluated through a spatial accessibility analysis, while the long-term economic outcomes are estimated by employing a quantitative spatial equilibrium model in order to simulate impacts on the regional commuting market. The short-term results indicate spatially highly uneven accessibility gains across amenities, depending on both the distribution of existing opportunities and municipal proximity to the highway. The long-term findings show a strong spatial concentration of impacts, with significant effects limited to a small part of the region. The model predicts that the affected area’s current population could be increased by up to 15 per cent through migration from other parts of the region, driven by improved access to the regional commuting market. Additionally, this migration could trigger an increase in residential floor space prices of up to 20 per cent. Together, the empirical results highlight the significantly localised and context-specific impact of transport interventions, which should be carefully considered by regional policymakers for the purpose of improving rural living standards efficiently.
By Yannic Prohaska; Vienna University of Economics and Business
Presented by: Yannic Prohaska, Vienna University of Economics and Business
Freight Hubs and the Geography of Production Abstract Trucking carries most freight value in the United States, but trade and economic‑geography models represent transportation costs as distance-based iceberg wedges. This paper argues that shipment-size–dependent trucking technologies generate trade costs with a network component: small consignments consolidate through hub‑and‑spoke less-than-truckload (LTL) terminals, while large truckload (TL) shipments move point‑to‑point. Hub routing adds detours and terminal handling that are largely orthogonal to bilateral distance, so delivered input prices vary systematically with proximity to consolidation hubs. I document these patterns using posted freight-rate quotes from a major online rate-aggregation platform. I compile standardized TL and LTL quotes across routes connecting 50 selected U.S. ZIP codes and construct measures of LTL hub proximity by mapping cross‑dock terminals from satellite imagery. In regressions that absorb origin and destination fixed effects, distance predicts rates in both markets, but hub proximity is an additional determinant for LTL: destinations closer to hubs face lower LTL rates, and the hub gradient is strongest on short routes where hub-and-spoke routing implies relatively large detours. To interpret the evidence, I model freight as a shipment-size–dependent technology with discrete TL/LTL choice and consolidation economies, and embed it in a quantitative spatial general‑equilibrium model with intermediate inputs. The model implies that hub proximity confers larger cost advantages for small-shipment firms; input sourcing amplifies (or partially offsets) these advantages through a “bad-neighbor/good-neighbor” channel; and the resulting manufacturing geography differs sharply from a distance-only benchmark. The framework links logistics infrastructure and hub placement to spatial policy, welfare, and path dependence in production geography.
By Ritabrata Bose; Arizona State University
Presented by: Ritabrata Bose, Arizona State University
Spatial Reach: A Structural Model of Multi-Margin Mobility Abstract Mobility decisions determine access to jobs, services, and social interactions, yet mobility behavior is typically modeled through observed outcomes such as distance traveled or trip frequency. We argue that these outcomes are equilibrium realizations of a deeper spatial decision. We introduce the concept of spatial reach: a latent scalar capturing the scale of economic and social engagement that individuals optimally choose given heterogeneous generalized mobility costs. Spatial reach represents the chosen extent of the accessible opportunity space and jointly governs participation, trip frequency, and realized travel distances. We develop a structural model in which individuals choose spatial reach at a medium-term horizon, taking into account nonlinear and individual-specific expected generalized cost (EGC) schedules that embed endogenous modal substitution. Conditional on reach, observable mobility outcomes emerge across multiple margins. This structure implies that policy-induced cost changes affect all behavioral margins through a single adjustment channel. The model is estimated using nationally representative French travel survey data combined with individual-specific cost schedules. We document strong diminishing returns to spatial expansion and substantial heterogeneity in purpose-specific elasticities. Notably, some activities expand with spatial scale while others consolidate, implying non-uniform responses to transport policies. Estimated spatial reach is heterogeneous across income groups and residential environments. Because participation, frequency, and distance are all governed by reach, transport policies primarily operate by contracting or expanding accessible opportunity space. Evaluations based solely on observed distance changes therefore understate behavioral responses and welfare effects. Spatial reach provides a unified and tractable framework for multi-margin mobility analysis and policy evaluation.
[slides]
By Nicolas Coulombel; Ecole des Ponts ParisTech    Guillaume Monchambert; University of Lyon
Presented by: Guillaume Monchambert, University of Lyon
Session 22: Managing Road Capacity: Lanes, Tolls, and Governance June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 11
Session Chair: André de Palma, CY Cergy Paris Université
Assessing the Impact of High Occupancy Vehicle Lane on Traffic Conditions: An Econometric Analysis of Policies on the Boulevard Périphérique of Paris Abstract This research studies the impact of the High Occupancy Vehicle Lane policy implemented on March 3rd, 2025, on the Boulevard Périphérique (motorway circling Paris) using traffic data from the city of Paris. A descriptive analysis shows that carpoolers experience a travel time reduction of about 23% compared to non-carpoolers on the non reserved lanes. To better understand how the infrastructure reacted to the lane reservation and get a causal analysis at a constant demand in the absence of a proper control group, econometrics is combined with traffic flow theory and simulation to understand technology changes in the road infrastructure and simulate counterfactuals. We find nuanced preliminary results for the policy. The maximum car demand the road can bear remains unchanged, but travel time per car is slightly increased. Travel times per passengers would be reduced only for very high levels of demand or if a strong modal shift were to be observed.
[slides]
By Martin Briand; ENPC, Institut Polytechnique de Paris, Univ Gustave Eiffel    Nicolas Coulombel; Ecole des Ponts ParisTech    Guillaume Monchambert; University of Lyon
Presented by: Martin Briand, ENPC, Institut Polytechnique de Paris, Univ Gustave Eiffel
Economics of High Occupancy Committed Lanes Abstract Carpooling is seen as a powerful demand management tool to tackle the problem of inefficient car usage. While some studies have shown that correct management of High Occupancy Vehicle Lane (HOVL) and High Occupancy Toll Lane (HOTL) can lead to social improvements, real world implementations provide mitigated results. It is indeed often argued that HOVLs fail to convince enough drivers to switch to carpool to be effective, and that HOTL reduce the attractiveness of carpooling while reinforcing inequalities among drivers. This paper proposes a new reserved lane management technique named High Occupancy Committed Lane (HOCL). We show what economics equilibria can arise from such a system and how it can be monitored to reach social optimum. We also prove that socio-economic value can be extracted through such a reserved lane management system in realistic cases. Finally, comparing the newly introduced High Occupancy Committed Lane (HOCL) system to more standard High Occupancy Vehicle Lane (HOVL), we prove that the former has a broader operation space than the latter, and is more efficient in terms of social cost.
[slides]
By Martin Briand; ENPC, Institut Polytechnique de Paris, Univ Gustave Eiffel    Nicolas Coulombel; Ecole des Ponts ParisTech    Guillaume Monchambert; University of Lyon
Presented by: Nicolas Coulombel, Ecole des Ponts ParisTech
Managing Road Capacity: Maintenance, Tolls, and Multi-Level Governance Abstract Road congestion and deteriorating infrastructure impose substantial economic and social costs, with estimates reaching up to 1% of GDP in highly congested economies. At the same time, public finances are increasingly constrained, and reliance on the "user-pays" principle has grown, prompting greater use of tolling schemes and private participation in financing, operating, and maintaining road networks. This paper examines the interaction between road maintenance, capacity, and pricing decisions in contexts where different operators share responsibilities. We analyze whether private maintenance and tolling strategies converge toward socially optimal outcomes, and under what conditions misalignments occur. Policy implications for optimal pricing, investment incentives, and the design of capacity and maintenance are discussed.
By André de Palma; CY Cergy Paris Université
Presented by: André de Palma, CY Cergy Paris Université
Session 23: Urban Air Quality and Vehicle Restrictions June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 12
Session Chair: Francis Ostermeijer, VinUniversity
The Equilibrium Effects of Urban Air Quality Policies: Application to the Grand Paris Low-Emission Zone Abstract Low-emission zones (LEZs) have become a central instrument to reduce urban air pollution. While existing evidence documents their effects on traffic and emissions, little is known about their general equilibrium impacts on urban spatial structure and welfare. This paper develops and quantifies a spatial general equilibrium model with endogenous commuting, transport mode choice and local air pollution. Extending standard quantitative spatial models of transportation, the framework incorporates mode choice and a spatially differentiated pollution process. Air quality emerges endogenously from commuting flows, residential and commercial activity and atmospheric dispersion, and feeds back into location decisions of households and firms. We apply the model to the Low-Emission Zone in the Grand Paris metropolitan area. The policy restricts internal combustion engine vehicles in a dense and highly connected region, effectively requiring electric vehicle ownership for car commuting within or into the regulated area. Using rich administrative and census microdata for Île-de-France, we estimate mode choice parameters, recover location elasticities from commuting flows and calibrate pollution dispersion to observed concentration patterns. Ignoring relocation, the LEZ generates large reductions in car traffic and emissions, with substantial air quality improvements. Allowing for full spatial adjustments significantly alters these results. Households and firms relocate to avoid constraints, reshaping employment geography and reducing direct exposure. These adjustments attenuate welfare costs through endogenous wage and rent responses but also dampen environmental gains. Overall, the results show that approaches which ignore spatial equilibrium responses tend to overestimate both the costs and the benefits of urban air quality policies.
By Eva Gossiaux; Paris School of Economics, Université Paris 1 Panthéon-Sorbonne    Mohamed Bahlali; AMSE
Presented by: Eva Gossiaux, Paris School of Economics, Université Paris 1 Panthéon-Sorbonne
Would nighttime deliveries be useful to reduce air pollution in cities? Econometrical evidence from Paris Abstract Taking advantage of a set of multimodal hourly traffic counters that differentiate between bikes, motorized two-wheelers, cars, light-duty vehicles, trucks and buses, we propose a new, data-driven, methodology to estimate the predicted impact of various off-peak truck delivery schemes. For this, we first estimate econometrically the impact of each type of vehicle’s hourly occupancy rate on traffic speed and on NOx and PM10 air pollution. Then, we use the estimated econometric coefficients to evaluate, after the rescheduling of trucks implied by each off-peak delivery scheme 1) the new traffic speed at each hour of each day, 2) the new occupancy rates implied by this new traffic speed, 3) the new NOx and PM10 concentrations at each hour of each day. We find that, even in the presence of a full rebound effect (i.e. if the reduction in truck traffic during the day is exactly offset by additional cars), air pollution is reduced. Eventually, we conduct a Cost-Benefit Analysis of the various off-peak delivery schemes and obtain that, in the absence of investments to reduce truck traffic noise at night, the costs of nighttime deliveries will exceed the benefits.
By Lucie Letrouit; Gustave Eiffel University
Presented by: Lucie Letrouit, Gustave Eiffel University
Engines of Change: Household and Mobility Impacts of Gasoline Motorbike Restrictions in Hanoi, Vietnam Abstract In July 2026, Hanoi will implement a limited zone restriction prohibiting gasoline motorbikes from entering Ring Road 1, its political and commercial core. Unlike prior studies of car-based restrictions, this policy targets two-wheelers, a mass-accessible mobility asset central to labor market access in Southeast Asia. Yet existing evidence focuses largely on aggregate traffic and pollution outcomes, leaving household-level welfare effects largely unexplored. We evaluate the policy using a difference-in-differences design comparing households inside and outside Ring Road 1 before and after implementation. The project combines baseline and follow-up household surveys (~2,000 respondents), a 2021 mobility benchmark (~25,000 respondents), six months of GPS mobility tracking, and geocoded air quality data. Surveys include income, debt, vehicle ownership, and an incentive-compatible willingness-to-pay module for electric motorbikes. The conference presentation will report preliminary analysis of baseline data, assess treatment-control comparability, and test identification assumptions, including balance and parallel trends, while documenting early electrification patterns and switching thresholds across income groups.
By Francis Ostermeijer; VinUniversity
Presented by: Francis Ostermeijer, VinUniversity
Session 24: Parking and Cruising: In Honor of Donald Shoup June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 13
Session Chair: Eren Inci, Sabanci University
Strategic Curb and Garage Pricing in Urban Parking Markets Abstract We develop a queueing-based Stackelberg model in which a public regulator sets the curb price anticipating the response of a profit-maximizing garage operator. Curb parking is prone to congestion, while garages provide nearly immediate access at a higher monetary or access cost. The analysis shows that curb pricing affects not only congestion but also private market power. A central and counterintuitive result is that underpricing curb parking can strengthen garage pricing power. When curb prices are low, cruising intensifies and guaranteed access becomes more valuable, enabling garages to raise prices and target high-valuation, time-sensitive drivers. Conversely, sufficiently high curb prices may reduce cruising and constrain garage markups. Over a range of parameters, increasing the public price reduces private prices and narrows the gap between curb and garage fees. The results may provide a theoretical explanation for the coexistence of low curb prices and high garage rates observed in many cities and highlight curb pricing as a strategic regulatory instrument, not merely a congestion-management tool.
By Amir Brudner; Hebrew University    Mehdi Nourinejad; York University    Opher Baron; University of Toronto
Presented by: Amir Brudner, Hebrew University
Cruising for Parking: the role of prices Abstract Pricing of parking is widely viewed as the first-best policy to reduce cruising, yet its empirical effectiveness is considered limited because parking demand is price-inelastic. We resolve this paradox by showing that cruising externalities are far more price sensitive than parking demand. For shopping streets in Amsterdam, we estimate a price elasticity of marginal cruising externality of -1.7, more than four times larger than the price elasticity of parking demand. Hence, even modest price increases can dramatically reduce cruising externalities. We further show that observed parking prices exceed the welfare-maximizing level implied by marginal external cruising cost pricing.
By Yue Cui; Vrije Universiteit Amsterdam    Jos van Ommeren; Vrije Universiteit Amsterdam
Presented by: Yue Cui, Vrije Universiteit Amsterdam
Parking in Motion: Autonomous Vehicle Loitering in an MFD City Abstract Autonomous vehicles do not make parking irrelevant. They make parking more tightly coupled with congestion, increasing the importance of pricing and curb management. We develop an integrated macroscopic fundamental diagram model of traffic congestion and parking, studying an HV benchmark, a fully AV environment, and mixed traffic during the transition between the two. With AVs, parking becomes endogenous because vehicles can enter the loitering flow rather than park. The model embeds parking search as a physical search-and-matching process within a macroscopic traffic framework, so matching efficiency depends on network conditions. Congestion slows parking search, longer parking search keeps vehicles in circulation, and additional circulation further increases congestion. The result is a superlinear congestion response: the maximum sustainable inflow can occur in the hypercongested region of the network. High inflow in an autonomous-vehicle city may, therefore, be sustained because vehicles spend less time parked and more time circulating, reducing the parked stock required to support a given inflow without improving the efficiency of the road network. Higher parking fees, insofar as they reduce the HV cruising-for-parking flow, may simultaneously make loitering relatively more attractive for AVs.
By Eren Inci; Sabanci University    Lewis Lehe; University of Illinois, Urbana-Champaign    Ayush Pandey; University of Illinois, Urbana-Champaign
Presented by: Eren Inci, Sabanci University
Session 25: Aviation Emissions: Policy and Regulation June 18, 2026 9:00 to 10:30 Location: Palazzo Baroni 14
Session Chair: Stef Proost, KULeuven
Pollution and congestion in airline networks: The role of emission charges Abstract This paper studies the two main externalities in air transportation, pollution and congestion, which are usually analyzed separately despite being strongly interconnected. We first develop a theoretical model in which airlines generate both externalities, showing that flight frequencies link the two. The model demonstrates that emission charges, unlike the largely theoretical congestion tolls, can act as a practical price-based mechanism that mitigates both pollution and congestion, while also incentivizing fleet renewal. We then test these predictions empirically by exploiting the EU Emissions Trading System (EU ETS) as an exogenous source of variation in emission charges. Using data on one-stop intercontinental routings from Europe, combined with congestion and alliance measures, we find that higher emission charges reduce frequencies, encourage the use of newer aircraft, and alleviate congestion. Moreover, we document a negative effect of alliances on frequencies and emissions, consistent with internalization among partner airlines. These results confirm that emission charges can function as an effective and comprehensive policy instrument, internalizing congestion in practice where tolls have failed.
By Ricardo Flores-Fillol; Universitat Rovira i Virgili
Presented by: Ricardo Flores-Fillol, Universitat Rovira i Virgili
Short-haul ban or zero-emission aircraft? Balancing social welfare and emissions control Abstract Recently, a global movement has taken various measures to address the high carbon emissions from the aviation industry. Among these, the short-haul ban policy prohibits short-haul flights in favor of more sustainable alternatives. However, it leads to the monopolization of high-speed rail (HSR). On the other hand, zero-emission aircraft, such as electric and hydrogen-powered planes, present a viable option that can maintain competition, although they are less effective in emissions reduction and more costly. In this study, we develop an economic model to compare these two approaches from the perspectives of social welfare and emissions within a hub-and-spoke network that includes one short-haul route and one long-haul route. Our findings indicate that the short-haul ban results in an increase in long-haul air traffic, but it simultaneously reduces Air-Air intermodal traffic connecting short and long-haul routes, which ultimately lowers overall long-haul air traffic and emissions. Compared to the short-haul ban, utilizing zero-emission aircraft is more favorable for passengers, although its emissions reduction effect is less pronounced. Interestingly, the emissions reduction effect of zero-emission aircraft increases with the aircraft's unit cost. By comparing social welfare across different scenarios, we propose optimal government strategies. Policymakers should consider the unit cost of carbon emissions and zero-emission aircraft. When the unit carbon emissions cost is high, the government favors the short-haul ban due to its significant emissions reduction effects. When the unit carbon emissions cost is low, no policy intervention is necessary. At an intermediate level, using zero-emission aircraft becomes socially optimal.
By Kun Wang; The Hong Kong Polytechnic University
Presented by: Kun Wang, The Hong Kong Polytechnic University
How to regulate the non-CO2 climate emissions from aviation Abstract The non-CO2 aviation emissions (mainly contrails) can, in total, be up to twice as damaging for climate as the CO2 emissions associated to the use of kerosene. As the warming effect of the non-CO2 emissions depends strongly on metereological conditions, it is difficult to attribute the non-CO2 warming effect to a particular flight. Without detailed meteorological information per fight only a blunt emission tax system that taxes all kerosene used in aviation at two to three times the current damage estimate of CO2 could work. Even if this can be justified from an environmental economics point of view, this risks to be unacceptable for the aviation sector. The result is that there is, at present, no active non-CO2 policy. The aviation industry has proposed a 20 year plan to improve the monitoring of these emissions before taking action to address the non-CO2 emissions. We propose a multi-stage regulation scheme to address this problem much faster. In the first stage, airlines and public sector agencies are subsidized to improve measurements. These measurement inputs are used to construct a contrail forecasting model. In the second stage, the model is used to propose alternative flight-paths. Airlines are incentivized to adopt the new flightpaths by subsidies that cover the additional flight operation costs. A numerical illustration for the wider EU-region shows that the mechanism proposed can lead to significant climate emission savings that are larger than the savings of CO2 emissions that result from the introduction of ETS or Sustainable Aviation Fuels for aviation.
By Stef Proost; KULeuven
Presented by: Stef Proost, KULeuven
Session 26: Keynote Lecture: Joan Walker June 18, 2026 11:00 to 12:15 Location: Aula Magna
Session Chair: Jonathan Hall, University of Alabama
Evidence Standards and Open Infrastructure for Travel Behavior Modeling in the AI Era Abstract Behavioral modeling in transportation faces an evaluation crisis at two scales. At the individual choice level, model specifications producing equivalent statistical fit can result in divergent policy recommendations, yet the field lacks shared infrastructure to make these differences visible and their lessons accumulable at scale. At the society level, behavioral forecasts and simulations for policy interventions are evaluated through informal plausibility checks and heuristics. At both scales, the field has many evaluation techniques but no shared standard for organizing them. This problem is about to intensify as LLM-powered simulations enter travel behavior modeling, producing seemingly plausible outputs of unclear behavioral fidelity. This talk presents two responses to this crisis. The first is an evidence ladder for organizing the evidence behind a model's predictive claims. The second is a prototype for the infrastructure at the discrete-choice level to make the policy consequences of specification choices visible. Together they give researchers and practitioners a shared standard for head-to-head model comparison, and the practices needed to accumulate evidence across studies. Building this infrastructure now is the field's urgent task before the AI momentum pushes the field’s fragmentation past the point of repair.
By Joan Walker; UC Berkeley
Presented by: Joan Walker, UC Berkeley
Session 27: Rail Investment, Capacity, and Welfare June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 8
Session Chair: Francesca Pagliara, University of Naples Federico II
A theory of optimal investment timing in railways Abstract In a random utility model for mode choice in transportation we allow for stochastic shifts of preferences following the upgrade of an existing mode. We take the measurable part of conditional indirect utility function on the upgraded mode to be scaled-up relative to that of a competing mode according to a geometric Brownian motion. Next the likelihood that one mode is preferred to the other is determined with logit and the marginal utility of income is constant. We identify the law of motion of both the likelihood that the upgraded mode will be chosen and the corresponding volume of the demand. We take the respective mode to be a railway, which is upgraded from traditional to high-speed railway. Our results reveal that the law of motion of the volume of the demand for HSR generalizes that used by public officials to estimate demand growth, and which are fully based on macroeconomic trends. The law of motion reduces to that adopted in practice if and only if the demand for competing modes (HSR and highway or airway) is expected to change in the same proportion over time. A good approximation of the discounted expected consumer surplus exists and can be used for the analysis of optimal investment timing on HSR. The latter depends on how optimistic are the forecasts about the total demand and the diverted demand from the competing mode. On two specific projects, we simulate our results based on available data to show under which conditions decisions were optimal.
By Asma Ouerhani; UNICAEN, CREM    Daniel Danau; Unicean, CREM
Presented by: Asma Ouerhani, UNICAEN, CREM
Allocating Rail Infrastructure Capacity on a Congested Shared Line: A Welfare-Optimizing Framework Abstract Achieving the European Union’s decarbonisation objectives requires a substantial modal shift from road and air to rail, implying a significant increase in both passenger and freight traffic. However, large parts of the European rail network are already congested, making infrastructure capacity a scarce resource, particularly on mixed-traffic lines. This raises a key regulatory question: how should limited railway capacity be partitioned across heterogeneous services to maximise social welfare? This paper develops a microeconomic framework to determine the optimal strategic partitioning of railway capacity on congested shared infrastructure. We model three downstream market segments (long-distance passenger services, regional public service obligations, and freight services) characterised by distinct competitive structures. Demand is represented through linear inverse demand functions and generalised cost functions incorporating frequency effects and modal competition. When aggregate capacity requests exceed available infrastructure, capacity becomes a binding physical constraint and maximum frequencies become decision variables of the infrastructure manager. We formulate a bi-level optimisation problem in which the infrastructure manager maximises total social welfare, anticipating endogenous pricing and frequency decisions in each downstream market. By explicitly modelling capacity as a binding constraint rather than a marginal congestion externality, the framework aligns railway capacity allocation with the airport slot allocation literature while extending existing rail models to three heterogeneous market segments. The model generates comparative statics on the welfare-optimal allocation of capacity and provides a decision-support tool for infrastructure managers under evolving European regulatory requirements.
By Chiara Widua    Francois Combes; IFSTTAR    Martin Koning; Université Gustave Eiffel
Presented by: Chiara Widua,
High-Speed Rail as a Strategic Transport Investment: Implications for Passengers’ Subjective Well-Being Abstract The relationship between high-speed rail (HSR) and subjective well-being (SWB) is becoming an increasingly significant area of research. The objective of this paper is to develop a theory-based, individual-level analytical framework that explicitly captures the mechanisms through which high-speed rail (HSR) affects subjective well-being. Using a Structural Equation Modelling (SEM) approach, the research represents HSR impacts as a network of relationships among latent constructs, with particular emphasis on service quality and its influence on well-being through key experiential and cognitive mediators, specifically, travel-related affect and cognitive evaluation. The analysis is further extended to explore how improvements in well-being translate into behavioural and business-related outcomes, such as customer loyalty and willingness to pay, thereby linking transport research with perspectives from management and service evaluation. Through this integrated framework, the paper offers three primary contributions. Firstly, it advances the transport and well-being literature by moving beyond aggregate-level analyses to focus on individual perceptions and experiences, enabling a more refined understanding of how infrastructure shapes everyday life. Secondly, it provides empirical evidence on the mediating role of travel experiences and life-integration factors, which have received limited attention in prior research. Thirdly, it connects well-being outcomes to behavioural responses relevant to transport operators and policymakers, offering a more comprehensive basis for assessing the broader impacts of high-speed rail investments.
By Francesca Pagliara; University of Naples Federico II    Alireza Moradpour; University of Naples Federico II    Tahseen Bashir; University of Naples Federico II
Presented by: Francesca Pagliara, University of Naples Federico II
Session 28: Transport Appraisal, Planning, and Efficiency June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 9
Session Chair: Andrew Smith, Institute for Transport Studies, University of Leeds
National Transport Planning under climate constraints: A methodological framework for Norway Abstract National transport planning is traditionally based on a predict-and-provide paradigm, which is increasingly misaligned with binding climate targets. In Norway, the statutory objective of a zero-emission transport sector by 2050 requires a shift towards vision-led planning, where strategies are evaluated for consistency with long-term climate constraints. This paper develops a methodological framework for operationalising such planning within the Norwegian National Transport Plan. We conceptualise the task as maximising social welfare from the transport system subject to a binding zero-emission constraint in 2050. The core contribution is an eight-step process that bridges qualitative backcasting and quantitative modelling, translating high-level goals into internally consistent, model-implementable policy packages. The framework soft-links passenger and freight transport models with the regional macroeconomic general equilibrium model NOREG2, enabling feedback between transport demand, modal choice, technology adoption, and economic structure. The framework is applied to six stylised pathways for 2050, including a frozen-policy reference, an EU-policy path, and four example strategies. Results show logically consistent and policy-relevant differences across strategies, particularly in cost distribution and incentives for Avoid–Shift adaptations. The study demonstrates that established transport models can support vision-led planning when embedded in structured backcasting and soft-linked modelling, offering a practical approach for climate-constrained national transport planning.
[slides]
By Paal Wangsness; Institute of Transport Economics    Wiljar Hansen; Intitute of Transport Economics    Inger Beate Hovi; Institute of Transport Economics    Bjørn Johansen; Institute of Transport Economics    Niels Kristensen; Insitute for Transport Economics    Anne Madslien; Institute of Transport Economics    Daniel Pinchasik; Institute of Transport Economics    Christian Steinsland; Institute of Transport Economics, Oslo    Askill H. Halse; Institute of Transport Economics
Presented by: Paal Wangsness, Institute of Transport Economics
Rethinking welfare appraisal for transport and multiple-market interventions with land use change Abstract The ‘rule-of-a-half’ benefit measure often used in transport appraisal (Tressider, 1968) omits certain benefits when land use changes. This research aims to provide a revised, robust method for welfare appraisals under land use change, not only for transport projects but for interventions including transport and planned land development (such as housing and commercial space). The paper includes key findings of a literature review and theoretical development work building on: Neuburger (1971); Bates (2006); Geurs et al (2010); Börjesson et al (2014); and Johansson (2021) among other sources. The work brings location attractiveness within the appraisal, addresses the appraisal of accessibility changes, and links the revised methods with modelling practice – including transport models and Supplementary Economic Modelling (SEM). The research has been undertaken on behalf of the UK Department for Transport.
By John Nellthorp; University of Leeds    Richard Batley; Leeds University    James Laird; Peak Economics Ltd    John Bates; John Bates Services    David Simmonds; Allanfield Consulting    Iven Stead; Department for Transport (UK)
Presented by: John Nellthorp, University of Leeds
Incorporating intermediate outputs into SFA using an equation system approach with application to local highways Abstract Stochastic Frontier Analysis is widely used to assess the performance of decision‑making units. But conventional models assume a single‑stage process in which inputs are transformed directly into final outputs. This ignores the role of intermediate outputs and obscures where inefficiencies arise within multi‑stage systems. Many production processes, especially in infrastructure industries, operate in two stages: the first generates intermediate outputs, which are then used to produce final outputs in the second stage. While multi‑stage frameworks have been developed in Data Envelopment Analysis, similar progress in SFA has been limited. Existing network‑type SFA models are often complex and do not yield measures of overall efficiency. This study introduces a Network SFA approach that incorporates intermediate outputs to better identify the sources of inefficiency. The model estimates both stage‑level and overall efficiency, allows analysis of returns to scale at each stage, and accommodates different levels of data aggregation, making it suitable when input data are aggregated but output data are available by division. A simplified multi‑step estimation procedure is provided along with authors’ written Stata package network SFA. The model is then applied to a two‑stage road maintenance process to demonstrate its advantages over single‑stage analysis. Key findings on returns to scale and inefficiencies at the different stages are reported and discussed, with implications for cost optimisation in Local Highway Authorities (LHAs) .
By Trang Tran; Institute for Transport Studies, University of Leeds    Phill Wheat; University of Leeds    Alex Stead; University of Leeds    Andrew Smith; Institute for Transport Studies, University of Leeds
Presented by: Andrew Smith, Institute for Transport Studies, University of Leeds
Session 29: Active Mobility: Cycling, Bike Sharing, and Infrastructure June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 10
Session Chair: Elena Maggi, Università dell'Insubria
Measuring the resilience of bike-sharing systems under different disruptions: counterfactual evidence from five European cities Abstract Urban transport systems are increasingly exposed to climate hazards, infrastructure failures, and public transport disruptions, yet evidence on how alternative modes absorb demand shocks remains fragmented across single-city and single-disruption studies. This study addresses this gap through a harmonised multi-city, multi-disruption counterfactual analysis of bike-sharing demand responses to acute urban disruption. The empirical setting covers Storm Boris flooding in Brno and Ostrava, the April 2025 Iberian Peninsula blackout in Badalona and León, and three public transport strike episodes in Frankfurt during February–March 2025. High-frequency trip data retrieved from operator APIs were validated, cleaned, and aggregated into daily demand series. Bayesian structural time-series models estimated counterfactual demand in the absence of disruption, while behavioural heterogeneity was assessed across peak-period and trip-duration segments. The results identify two contrasting resilience mechanisms: demand suppression under direct physical exposure and modal substitution when competing transport modes fail. Flooding reduced daily bike-sharing demand by 22% in Brno and 34% in Ostrava, with losses concentrated in off-peak and medium-to-long trips. The blackout was followed by post-event demand increases of 17% in Badalona and 32% in León, although behavioural composition differed between the two cities. Frankfurt strikes produced the clearest substitution effect, increasing daily demand by 14–17% and morning-peak demand by 37–65%, while off-peak demand remained statistically unchanged. The findings show that bike-sharing resilience is not a universal system attribute, but a conditional behavioural response shaped by disruption type, temporal segment, and trip-length profile, with implications for differentiated urban contingency planning.
By Sorath Shah; Masaryk University    Michal Dziecielski; Adam Mickiewicz University
Presented by: Sorath Shah, Masaryk University
Safe lanes, cycling students. Infrastructure investment for sustainable commuting by bike in Italian universities Abstract Commuting to university is a puzzling topic when living at a short distance from college. Whereas pro-ecological attitudes might call for unimodal cycling to connect to college sites, concerns about road safety may impede shifts to low-impact habits and favour motorized vehicles, resulting in dissonance habits. Although investments in safe urban cycling infrastructure are critical for achieving sustainable mobility goals, yet evidence on their effectiveness in changing commuting behaviour remains limited. This paper examines how cycling infrastructure quality and perceived safety mediate the relationship between pro-environmental attitudes and modal shift among university commuters in Italy. Using data from 51 Italian universities, we employ structural equation modelling to assess infrastructure-behaviour linkages. Relying on self-determination theory constructs, results show that safe cycling infrastructure is a necessary condition for translating social pro-environmental norms into behavioural change, while intrinsic environmental motivations can partially substitute for infrastructure quality. Policy implications suggest that public investment in dedicated cycling infrastructure yields heterogeneous returns depending on the target population's motivational profile, requiring differentiated policy approaches for Italian university cities.
By Daniele Crotti; University of Insubria    Elena Maggi; Università dell'Insubria    Jurgena Myftiu; University of Bergamo
Presented by: Elena Maggi, Università dell'Insubria
Session 30: Behavioral Responses to Road and Fuel Prices June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 11
Session Chair: Alice Ciccone, Institute of Transport Economics
The salience of toll ring prices Abstract Standard economic theory assumes that individuals fully optimize in response to taxes and price instruments. However, limited attention and cognitive constraints imply that non-salient prices may not be fully incorporated into decision-making. While prior research shows that increasing tax salience affects behavior, causal evidence remains limited in the context of road tolls. This paper examines whether increasing the salience of road toll prices and costs affects driving behavior and attitudes toward tolls. We conduct a large-scale field experiment in the Oslo area in partnership with the toll service provider Autosync. More than 100,000 customers form the study population, of which 6,000 toll tag owners are randomly assigned to receive weekly emails detailing toll prices and their individual toll expenditures. Administrative data on weekly toll passings allow us to estimate behavioral responses, while survey data measure knowledge and attitudes. The study provides clean experimental evidence on whether a low-cost informational intervention can influence routine travel decisions in a high-frequency setting. Beyond behavioral effects, we examine whether increased salience alters public support for toll policies. The results inform both the literature on tax salience and the design of effective and politically sustainable transport demand-management policies.
By Goeril Andreassen; INSTITUTE OF TRANSPORT ECONOMICS    Alice Ciccone; Institute of Transport Economics    Askill H. Halse; Institute of Transport Economics    Karen Hauge; Frisch Centre
Presented by: Goeril Andreassen, INSTITUTE OF TRANSPORT ECONOMICS
The Price of Speed: Estimating Intensive-Margin Responses to Fuel Prices on French Toll Motorways Abstract Fuel prices affect road transport not only by changing how much people travel, but also by changing how they drive. This paper studies this short-run intensive margin on French toll motorways, focusing on travel time per kilometre as a measure of speed adjustment. The main empirical challenge is that observed motorway speeds also reflect traffic conditions. When fuel prices change, traffic volumes and the mix of light and heavy vehicles may change as well, making it difficult to distinguish the behavioural channel of interest from traffic-state effects. To address this issue, the analysis uses six-minute loop-detector data from French toll motorways and estimates a composition-aware reduced-form specification. The model distinguishes light- and heavy-vehicle fuel-price channels and conditions on traffic state, vehicle composition, weather, calendar disruptions, and sensor-specific diurnal patterns. The paper contributes a framework for estimating conditional intensive-margin responses to fuel prices in a setting where traffic load, vehicle mix, and driving speed are jointly observed.
By Julian Barbella; ENPC-IP Paris    Nicolas Coulombel; Ecole des Ponts ParisTech    Guillaume Monchambert; University of Lyon
Presented by: Julian Barbella, ENPC-IP Paris
Road Pricing: Travel Behavior and Public Support Abstract Using a large-scale randomized field experiment, we study behavioral responses to and public support for Pigouvian road pricing. Leveraging financial incentives, we find that drivers subject to distance-based charges reflecting external costs of driving—varying by location, time, and vehicle type—reduce their driving-related externalities by 5.3%, corresponding to a price elasticity between –0.07 and –0.15, with much weaker effects among users of battery-electric vehicles (BEVs). Furthermore, providing information on the expected benefits of road pricing enhances public support for such policies, whereas experience with road pricing has little impact.
By Alice Ciccone; Institute of Transport Economics    Cloe Garnache; Oslo Metropolitan University
Presented by: Alice Ciccone, Institute of Transport Economics
Session 31: Value of Time and Demand Model Identification June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 12
Session Chair: Thijs Dekker, Institute for Transport Studies
What Determines Gender Differences in Value of Time? The Impacts of Residential and Work Location Choices Abstract We examine how residential and work locations affect gender differences in time-use burden among ex-ante homogeneous couples in a city. We theoretically derive gendered values of time (VOTs) by location in spatial equilibrium. We show that (i) changes in VOT according to residential and work locations are determined by shadow prices of time and budget, (ii) households residing closer to the central business district (CBD) have higher VOTs, and (iii) housework division varies by location, and VOTs differ between spouses spatially. Simulations substantiate these findings. Furthermore, our stated preference experiment confirms that VOT varies with residential location, after controlling for gender and household characteristics.
By Ashley Wan-Tzu Lo; Tohoku University    Tatsuhito Kono; Tohoku University
Presented by: Ashley Wan-Tzu Lo, Tohoku University
Omitted variable bias in testing of theoretical properties of demand models and econometric solutions Abstract This paper addresses an issue which has not been mentioned in the literature on demand model testing: in the presence of omitted variable bias (OVB), the theoretical testing of demand functions must account for the impact of correlated but omitted variables. This is required because theoretical frameworks postulate conditions for the true value of the parameters and not for the value of the estimated parameter, which may be inclusive of the effect of omitted variables. Instrumental variables were used to control for omitted variable bias in the price of transport, the price of non-transport goods and income. Using data from the National Travel Survey in England between 2002 and 2019, and focusing on three specific theoretical tests: homogeneity of prices and income, transport being a normal good and transport not being a Giffen or Veblen good, it concludes that the results of the tests change depending on whether omitted variable bias is corrected for, with those corrected for OVB being the correct tests. This means that, when theoretical properties are tested for, one should correct for the impact of OVB as otherwise they may be misled by the results. The size of the effect of omitted variable bias was calculated, including its standard errors. When constraints are applied so that the homogeneity property is met by design, some of the parameter estimates may be different compared to a constrained regression without consideration for OVB. The magnitude of the difference will depend on the size of the effect of OVB.
By Diogo Magriço; University of Leeds    Richard Batley    Manuel Ojeda-Cabral; ITS Leeds
Presented by: Diogo Magriço, University of Leeds
How robust are Value of Travel Time estimates in simple time-cost trade-offs? Thijs Dekker – Institute for Transport Studies Abstract In this paper I examine the robustness of value of travel time welfare estimates from simple time-cost trade-offs used in stated preference and revealed preference studies and argue that a lot of the variation in derived VTT estimates is the result of inherent uncertainty (i.e. limited identifiability) in the underlying data, and hence the observed choices can support a wide range of potential VTT values. This is examined in the context of Kang and Vasserman (2025) and illustrated using the 2024 Dutch national VTT study. Potential ways forward are identified both in terms of econometric analysis and experimental design.
By Thijs Dekker; Institute for Transport Studies
Presented by: Thijs Dekker, Institute for Transport Studies
Session 32: Air–Rail Connectivity and Passenger Choice June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 13
Session Chair: Martin Adler, SEO Amsterdam Eocnomics
Determinants of Air-Rail Intermodality: Evidence from the 2024 French National Air Passenger Survey. Abstract As efforts to reduce aviation emissions intensify, air–rail intermodality is increasingly promoted as a greener substitute for short-haul feeder flights. However, in France, integrated single-ticket intermodal products are relatively new, despite qualitative evidence that passengers already build their own intermodal itineraries. Our work aims to analyze both the prevalence and the determinants of such intermodal connections among air passengers. We exploit a large survey data from the french Civil Aviation Authority. To approach the question of intermodality, we aim to evaluate the behaviour of passengers who have the choice between train, flight and road on one section of their journey. To do so, we select passengers departing from 4 regions: Nouvelle-Aquitaine, Occitanie, Provence-Alpes-Côte d'Azur, and Auvergne-Rhône-Alpes; and taking an international flight in Paris Charles De Gaulle (CDG) aiport. The first results of the multinomial logit model show that several characteristics affect the intermodal choice, among which: the purpose of travel, and the number of people travelling together.
By Sai BRAVO    Zélie Gankon; ENAC    IMANE HOUMMIRAT; Université de Montpellier    Estelle MALAVOLTI; enac & TSE    Antoine Ronsain; ENAC
Presented by: Zélie Gankon, ENAC
Analysis of Route Choice for Passengers Traveling between North America and Southeast Asia Abstract The air passenger market between North America and Southeast Asia is a critical ultra-long-haul segment characterized by intense competition between direct services and connecting flights via East Asian hubs. This study clarifies the factors influencing passenger route choice by incorporating both direct and connecting options into a single analytical framework, addressing a gap in existing literature that often excludes direct flights. Utilizing a nested logit model, the research analyzes travel patterns between six major Southeast Asian cities and five North American hubs. The model evaluates passenger decision-making based on three primary variables: average fare, total flight time, and weekly flight frequency. The empirical results demonstrate that while passengers predictably favor lower fares, higher frequencies, and reduced travel times, there is a remarkably strong preference for direct flights, significant at the 1% level. Notably, this preference persists even when controlling for total flight time, highlighting a distinct aversion to the transfer process itself. This finding indicates that for passengers, the convenience of a non-stop flight extends beyond mere time savings. Consequently, airlines providing connecting services through East Asian hubs face significant pressure to compete with the growing availability of direct services, necessitating strategic adjustments in pricing and scheduling to attract passengers who strongly prefer avoiding transfers.
By Yu Morimoto; Konan University
Presented by: Yu Morimoto, Konan University
The economic value of air and rail connectivity: Network scope and quality Abstract Air and rail infrastructure provide critical access to urban economies and shape the connectivity between functional urban areas. This paper proposes a novel set of eigenvector-based indicators to evaluate airport networks in conjunction with rail connectivity. The indicators capture both network scope and connection quality by incorporating economically relevant linkages between pairs of functional urban areas. They measure the performance of individual network nodes in providing access to relevant markets. Our results show that the highest-performing nodes provide direct access to nearly half of their economically relevant links. By contrast, despite performance being a function of realized economic links, medium-sized nodes exhibit substantially lower performances of usually well below 25%. The proposed performance metrics are highly sensitive to the inclusion of modal alternatives, functional form, urban clustering, and threshold values, even among nodes with similar characteristics. The findings highlight both the potential and the limitations of connectivity-based performance metrics for policymakers and infrastructure providers.
By Martin Adler; SEO Amsterdam Eocnomics    Christiaan Behrens; VU University
Presented by: Martin Adler, SEO Amsterdam Eocnomics
Session 33: Airline Profitability and Performance June 18, 2026 13:30 to 15:00 Location: Palazzo Baroni 14
Session Chair: James Odeck, Norwegian University of Science and Technology
The airline profits paradox Abstract The airline profits paradox Airline economic profits have rarely been positive for the industry in aggregate worldwide, or for most airlines. Shareholder value has been consistently destroyed over decades. Yet most airline city-pair markets are monopolies or oligopolies, for which standard theory would predict positive, if not high, economic profits. Moreover, capital continues to be allocated to further expand the airline industry, whereas theory would anticipate capital flight. How can this apparent paradox, of economic losses yet oligopolistic structures and continued expansion, be explained? We survey research on the application of oligopoly theory to airline markets, on the effect of government intervention through subsidy, state-ownership, state-aid and bankruptcy regimes, on empty core theory, on product differentiation, and on business environment factors. We find that the high markups over marginal costs predicted by oligopoly models can be consistent with economic losses in the presence of the industry’s extensive fixed costs, low marginal costs and perishability of the seat product, in the short run. But this cannot explain persistent losses in the long run. Empty core theory may explain market instability, but not persistent economic losses. Government intervention through subsidies and state ownership, and supportive bankruptcy regimes preventing the exit of unprofitable capacity are found to be part of the explanation of this paradox. The literature from business strategy offers some insight into, but fails to explain its confluence with a continued inflow of capital. While different theories show light on the paradox there is still scope for further research.
By Hans-Martin Niemeier; University of Applied Sciences Bremen
Presented by: Hans-Martin Niemeier, University of Applied Sciences Bremen
Service Continuity in U.S. Regional Air Markets During and After the COVID-19 Pandemic Abstract Since the COVID-19 pandemic, U.S. regional airlines have faced a dual challenge of pilot attrition and rising labor costs, leading to a substantial reduction in flight operations. Between 2019 and 2024, U.S. regional airlines reduced their domestic departures by nearly 24%, accompanied by a 19% contraction in passenger enplanements. Over the same period, mainline carriers adjusted their reliance on regional partners, increasing regional operations during the pandemic but scaling them back in the post-pandemic recovery. Against this backdrop, our paper examines determinants of route survival for U.S. regional airlines during and after the pandemic. Using airline-route panel data covering the 2017 – 2024 period, we estimate Cox proportional hazards models to analyze route discontinuation risk. The results show that routes operated by independent regional airlines are more likely to be discontinued than those operated by wholly-owned subsidiaries of mainline carriers. Routes serving large hub airports exhibit higher survival probabilities, reflecting the importance of network connectivity beyond local markets. Support through the U.S. Essential Air Service (EAS) program also improves route survival. The analysis controls for market demand conditions, including population and per capita income, as well as route-level characteristics such as load factor, passenger yield, unit cost, stage length, and aircraft size, capturing the adaptive strategies regional airlines have implemented in response to the pandemic period challenges. This study contributes to the literature by providing the first post-pandemic empirical evidence on how ownership status, network position, and policy support jointly shape route survival and network resilience in U.S. regional airline operations.
By Richard Klophaus; Worms University of Applied Sciences
Presented by: Richard Klophaus, Worms University of Applied Sciences
Performance Trajectories of European Airlines Through COVID-19: A Stochastic Frontier and Malmquist Index Approach Abstract This study examines the efficiency and productivity performance of 19 European airlines over the period 2013-2024, covering the pre-COVID-19 years, during the pandemic, and the early post-pandemic recovery. A panel Stochastic Frontier Analysis (SFA) with time-varying technical efficiency is combined with the Malmquist Productivity Index (MPI) to evaluate how airlines adjusted their production behavior across these distinct phases. To assess functional-form robustness, we estimate a translog stochastic frontier and allow for business-model heterogeneity through low-cost carrier (LCC)-specific interactions. A central methodological contribution concerns the treatment of technological heterogeneity: the LCC indicator is incorporated directly into the production frontier rather than the inefficiency term, enabling separation of structural technological advantages from managerial inefficiency. In addition, average stage length is included in the inefficiency equation to capture network-structure differences across airlines, thereby distinguishing pure technical inefficiency from performance variation driven by route configuration. Frontier estimates indicate that labor, fleet size, and block minutes exert significant positive effects on output, with utilization (block minutes) showing the strongest impact. The LCC coefficient is positive and significant, suggesting structural advantages in production technology. Year effects reveal a pronounced frontier shift downward in 2020-2021, followed by recovery during 2022-2024. Technical efficiency declined sharply during the pandemic but returned to pre-crisis levels thereafter. Non-parametric tests confirm that the pandemic years differ significantly from both pre- and post-COVID periods. MPI results show moderate pre-pandemic growth, a substantial contraction in 2020-2021, and efficiency-driven rebound subsequently.
By Tassew Tolcha; Molde University College    James Odeck; Norwegian University of Science and Technology
Presented by: James Odeck, Norwegian University of Science and Technology
Session 34: Freight Carriers: Market Structure and Pricing June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 8
Session Chair: Anne Lange,
Congestion and the market structure of fleets Abstract Fleets are becoming increasingly prevalent in urban transport. Unlike traditional vehicles, fleet operators may be able to internalize, at least partly, the congestion cost they impose on other fleet vehicles, thereby reducing the level of congestion. We analyze the optimal market structure in the market for fleets by comparing welfare under a monopolistic and a competitive fleet. Our model exhibits a unit mass of heterogeneous travelers choosing between private vehicles, vehicles belonging to a fleet, and public transport. We assume that private and fleet vehicles create congestion, while public transport does not. We emphasize a stark trade-off: a monopolistic fleet may be better at internalizing the congestion cost, but, as standard, may inefficiently reduce the amount of fleet vehicles dispatched. We find that, when the cost of a private car is high (such as in the case of dense urban contexts), and travelers choose between fleet vehicles and public transport only, a monopolistic fleet may be welfare superior to a competitive fleet, due to its superior ability to internalize congestion costs. In situations where the cost of using a private car is low or public transport options are absent, travelers face a choice between private cars and fleet vehicles only. Under these conditions, a competitive fleet structure is welfare-superior to a monopolistic one, as cost-reflective pricing, fostered by competition, exerts a greater influence on welfare than the internalization of congestion externalities.
By Federico Boffa; Libera Università di Bolzano    Mohamed Ceesay; University of Firenze    Alberto Iozzi; Università di Roma Tor Vergata
Presented by: Federico Boffa, Libera Università di Bolzano
Booking-Price Elasticity and Cost Pass-Through in Digital Parcel Brokerage: Evidence from a Multi-Carrier Platform Abstract Digital parcel brokers act as intermediaries between carriers and customers: they purchase transport capacity from multiple carriers and offer customers an all-in shipping price based on carrier-specific procurement costs. This paper studies two core objects in pricing in a platform setting: (i) booking-price elasticities for standardized parcel services and (ii) procurement-cost pass-through into customer-facing prices. We integrate multiple internal databases to link each shipment to the customer-facing all-in price, the corresponding carrier charge, and detailed booking attributes. The final sample contains more than 500,000 shipments over a multi-year period across multiple European origin markets and includes shipments executed by a wide set of carriers. A central identification challenge is endogenous pricing: quoted prices include markups that may respond to information correlated with unobserved demand conditions, causing OLS estimates to confound demand effects with brokers’ pricing decisions. The analysis shows that both booking responsiveness and cost pass-through vary across origin markets and shipment segments, implying that uniform platform-wide pricing and pass-through rules are not granular enough. The findings provide evidence on demand and cost incidence in digitally mediated logistics markets and inform market–segment calibration of pricing and margin policies.
By Kristof Horvat; University Of Luxembourg    Anne Lange
Presented by: Anne Lange,
Session 35: Public Transit in Developing-Country Contexts June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 9
Session Chair: Leif Soerensen, University of Applied Sciences and Arts (HAWK)
The Impact of Crime Perception on Public Transport Demand: Evidence from Six Latin American Capitals Abstract Public transportation systems are central to reducing traffic congestion and urban pollution, yet in high-crime cities, safety concerns may undermine their use. This paper studies how crime affects the demand for public transportation, using three pre-registered experiments with over 5,000 participants across six Latin American capital cities. First, we find that users place a high value on safety in public transportation: reducing crime is valued at over 50% of the fare. Second, crime reduces the likelihood of choosing public transportation over private alternatives, especially among women. For some users, even eliminating fares does not offset the deterrent effect of higher crime. As a result, crime lowers the responsiveness of public transportation demand to fare changes, limiting the effectiveness of subsidies in increasing ridership in high-crime settings. Finally, we find that in a budget allocation task, safety concerns do not reduce support for environmental goals, suggesting room for coordinated interventions. Overall, our findings show that crime creates a negative externality on congestion and pollution by limiting shifts to cleaner modes of transportation. They underscore the importance of integrating public safety into transportation and environmental policy, and highlight the potential for aligning safety and sustainability objectives.
By Santiago De Martini; University of Southern California
Presented by: Santiago De Martini, University of Southern California
Accessibility under Weak Regulation in Saraburi, Thailand – A Gravity-Based Assessment of Decentralized Public Transport Provision Abstract This research examines spatial accessibility in Saraburi, Thailand, a secondary city where public transport services are privately operated under weak regulatory enforcement. In this institutional setting, route structures and service frequencies are largely determined by expected fare revenues rather than coordinated planning objectives. The study investigates how decentralized, profit-oriented routing decisions shape the spatial distribution of access to employment, education, health, civic, commercial, and leisure opportunities. Accessibility is modeled using a negative exponential gravity framework in which opportunities are discounted by generalized travel cost. The cost function differentiates walking, waiting, and in-vehicle time, incorporates fare costs through a locally benchmarked value of time, and includes a congestion penalty when estimated passenger load factors exceed vehicle capacity. Purpose-specific impedance parameters allow comparison across opportunity types within a unified behavioral structure. The empirical implementation combines digitized route networks, pedestrian accessibility, high-resolution population data, land-use-based employment proxies, and point-of-interest information. Preliminary results indicate threshold-like spatial patterns consistent with revenue-density constraints. Accessibility changes abruptly at the margins of commercially viable corridors, producing a service-or-no-service spatial structure. A substantial share of subdistricts lacks effective public transport access to several essential opportunity types. Moreover, accessibility to education and employment appears only weakly aligned spatially, suggesting that decentralized routing does not necessarily internalize complementarities between opportunity categories. The working paper is intended to contribute empirical evidence on how weak regulatory environments shape accessibility outcomes and will be extended by robustness checks and regional extensions for further (comparative) analysis.
By Thammarat Bunjaroenphornsuk; University of Göttingen    Jan Schlüter; HAWK Holzminden    Leif Soerensen; University of Applied Sciences and Arts (HAWK)
Presented by: Leif Soerensen, University of Applied Sciences and Arts (HAWK)
Session 36: Experimental Evidence on Toll Acceptability June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 11
Session Chair: Philippe Gagnepain, Paris School of Economics-Paris 1
Voting Between Social Justice and Rewarding Green Transportation Behaviour - A Laboratory Experiment Abstract While environmental taxation represents an effective and potentially essential tool for the environmental transition in the transportation sector, it may face public resistance. The experimental economic literature has already demonstrated how experiencing the benefits of taxation can influence tax acceptability. However, the impact of tax revenue recycling on acceptability has not been thoroughly explored. This study proposes a laboratory experiment in which residents of a city interact when choosing between individual commuting modes. The residents are heterogeneous, with half being advantaged by lower transportation costs. These choices generate congestion and pollution costs. Residents are asked to vote for taxes that address these externalities, with three different revenue recycling options: lump-sum redistribution, social cushioning and environmental redistribution. In this study, tax implementation shows only a minor impact on policy support. Taxes remain highly effective, regardless of the redistribution mechanisms implemented, indicating that participants internalise the externalities' costs and modify their behaviour accordingly. Some psycho-social constructs are shown to impact policy support. As expected, participants demonstrate a clear preference for the redistribution option that favours them. However, an order of preference is also found among options offering similar expected gains. The level of support for any given policy depends upon the set of policy alternatives with which it is being compared. This highlights the importance of awareness regarding the diverse range of potential policies, which can influence the perception of each policy. This paper presents preliminary findings which warrant further investigation.
By Tanguy Richard; IFP Energies Nouvelles    Benoit Cheze; IFPEN    Laurent DENANT-BOEMONT; Univ Rennes & CREM CNRS    Pascal Gastineau    Frédéric Martinez; Gustave Eiffel University
Presented by: Tanguy Richard, IFP Energies Nouvelles
An Experimental Perspective on Urban Toll Acceptability Abstract This paper investigates the effectiveness and acceptability of urban tolls as a policy instrument in transport economics to reduce private car use. It focuses on everyday commuting decisions under time constraints and heterogeneous spatial access to public transport infrastructure. We consider an experimental approach to test how commuters adjust their modal choices after road pricing is introduced and when congestion is endogenously formed. We also assess the social acceptability of tolls and the role of distributive fairness in shaping support. We compare alternative pricing structures to explore how differentiated charges affect both modal shift and perceived equity. Our results show that congestion pricing reduces car use and encourages substitution toward public transport when the infrastructure is available. However, public support remains limited among individuals who perceive the scheme as unfair. Our findings thus underscore the importance of integrating efficiency and equity considerations when designing politically feasible urban transport policies.
By Philippe Gagnepain; Paris School of Economics-Paris 1    Thibaut Lapeyre; Paris School of Economics-Paris 1    Carine Staropoli; Paris School of Economics
Presented by: Philippe Gagnepain, Paris School of Economics-Paris 1
Session 37: Information Frictions in Vehicle Use June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 12
Session Chair: Bruno De Borger, University of Antwerp
Usage-based Insurance and Rational Inattention in the Era of Connected Vehicles Abstract Usage-Based Insurance (UBI) leverages telematics to link premiums to a monitored driving effort signal, offering a mechanism to correct accident externalities through signal-contingent pricing. Standard principal-agent theory predicts that such monitoring mitigates moral hazard and improves welfare. However, these predictions rely on the assumption that drivers fully process risk-relevant information. This paper incorporates rational inattention into a principal-agent insurance framework to examine how information frictions alter the welfare implications of monitoring technology. Drivers in our model face costly information acquisition about uncertain driving conditions before selecting effort, generating behavioral wedges between perceived and actual risk. The insurer offers contracts comprising an upfront premium and a signal-contingent indemnity schedule, and optimizes against the agent's endogenous information strategy and effort choice. We derive analytical results for a tractable action space and validate them numerically for richer specifications. Our findings confirm that standard contract properties, such as indemnity schedules decreasing in signal quality, are preserved. However, we demonstrate that monitoring can reduce consumer surplus relative to an unmonitored benchmark when cognitive costs appear, as insurers design more aggressive incentive structures to overcome informational frictions. Despite this redistribution, insurer profits and aggregate welfare increase with monitoring, driven by improved risk pricing. These results carry direct policy implications: regulators evaluating UBI programs must account for surplus redistribution from policyholders to insurers and consider complementary information provision policies to safeguard consumer welfare.
By Sen Yan; New York University    Daniel Vignon; New York University
Presented by: Sen Yan, New York University
Information provision by navigation apps and local transport externalities Abstract We study information provision by navigation apps in a parallel network consisting of a major road and a local network. We distinguish between two types of app provided information: shortcuts and navigation quality. First, we show that, if providing cost-reducing information on shortcuts through the local network strongly increases the marginal external cost of traffic imposed on the local population, then it is socially optimal not to offer this type of information at all. Moreover, it is always optimal to provide more information that improves the quality of guidance through the local road network than information on shortcuts. Second, if congestion on the main road is important relative to the external costs on the local network the social optimum will imply more information on shortcuts than what the private operator would offer. Finally, we analyze the response of the authority responsible for the local network to the information provided by the app. We show that the local government will respond differently to different types of information. More specifically, it may react to better quality information by reducing traffic calming, while increasing traffic calming when the app provider offers more information on shortcuts. lower quality guidance through the local network.
By Bruno De Borger; University of Antwerp    Stefanie Peer; Vienna University of Economics & Business    Antonio Russo; Institut Mines-Télécom Business School
Presented by: Bruno De Borger, University of Antwerp
Session 38: Senior Drivers and License Renewal Policy June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 13
Session Chair: Tano Kannegiesser, TUD Dresden University of Technology
The Effect of Driver’s License Renewal Extensions on Fatal Crashes among Senior Drivers: Evidence from Maryland Across Urban and Rural Areas Abstract We examine the impact of Maryland’s driver’s license renewal period extension on fatal crashes involving senior drivers using county-level data from 2012–2019 and the Synthetic Difference-in-Differences method. The reform increased fatal crashes among older drivers, particularly in urban counties, with some estimates statistically significant. The findings suggest that renewal cycle extensions may exacerbate crash risks for senior drivers, especially in urban areas.
By Jae-Yu Jung; Michigan State Univerisity
Presented by: Jae-Yu Jung, Michigan State Univerisity
Cost-Benefit Analysis of Age-based Screening for Driving License Renewal Abstract As older car drivers are often seen as unsafe, several countries employ age-based screening (ABS). This road safety policy shortens the driving license validity based on age and makes renewal conditional on medical examinations. Since most scientific literature finds ABS ineffective at fulfilling its safety-enhancing purpose, its continued adoption is questionable. This paper is the first to perform a comprehensive Cost–Benefit Analysis of ABS, thereby improving policymakers' understanding of its welfare implications. Specifically, we investigate a hypothetical ABS introduction in Germany that mirrors common EU practice (screening from age 65 every 5 years; medical exam at a general practitioner; €25 fee). In a first step, we empirically derive two key input parameters: the number of driving licenses revoked annually and the resulting annual mileage reduction. Then, we use this information to quantify primary and secondary market effects of ABS by calculating changes in social opportunity cost, consumer surplus, transport externalities, and government revenue. Finally, we develop a strategy to discount the policy´s future benefits and costs to account for the fact that some impacts of ABS vary over time and have a dynamic nature due to specific age-cohort effects. Assuming no substantial intensive-margin safety benefits, our main result is a strongly negative net present value of $-$€605 million for our baseline implementation year 2024, equivalent to a welfare loss of around €12,151 per revoked driving license. The corresponding benefit-cost ratio is $-$4.6. The outcome is heavily driven by the loss of consumer surplus, as ABS prevents older drivers from choosing their preferred travel mode for the rest of their lives. While the negative welfare effect holds in 86% of cases in a Monte Carlo simulation, a structural scenario analysis reveals that the outcome could flip if real-world behavioral or safety mechanisms deviate drastically and optimistically from the empirical consensus. We conclude that the specific ABS model evaluated is highly inefficient from an economic standpoint. Although real-world licensing policies reflect broader considerations beyond economic efficiency, policymakers should reevaluate the use of general ABS based on contemporary scientific evidence.
By Tano Kannegiesser; TUD Dresden University of Technology    Stefan Tscharaktschiew; TUD Dresden University of Technology
Presented by: Tano Kannegiesser, TUD Dresden University of Technology
Session 39: Fleet and Network Emissions June 18, 2026 15:30 to 16:30 Location: Palazzo Baroni 14
Session Chair: Jan K. Brueckner, University of California, Irvine
Sustainable Network Planning: Assessing Potential Mitigation of Aviation Emissions through Centralized Network Redesign of Long-haul Connections Abstract Sustainable network planning strategies have the potential to mitigate aviation emissions. However, the extent to which emissions can be reduced using such strategies is not clear. This paper develops a centralized, welfare-oriented optimization framework to estimate an upper bound on the emission reduction that can be achieved restructuring current networks. The model integrates network design and fleet assignment decisions, explicitly investigating the trade-offs between environmental performance, economic viability, and passenger service quality. The model is applied to long-haul passenger flows between Europe and intercontinental destinations, testing the effects of hub reallocation and the rebalancing of connecting and nonstop services.
By Nicolò Avogadro; University of Bergamo    Sebastian Birolini; University of Bergamo    Paolo Malighetti; University of Bergamo    Stefano Paleari; University of Bergamo
Presented by: Nicolò Avogadro, University of Bergamo
Capital Flight? Do Airlines Move Fuel-Efficient Planes to High-Fuel-Tax Routes Abstract see uploaded file
By David Agrawal; University of California, Irvine    Jan K. Brueckner; University of California, Irvine    Quinton White; US Treasury
Presented by: Jan K. Brueckner, University of California, Irvine
Session 40: Transit Demand: Behavior and Competition June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 8
Session Chair: Takao Dantsuji, Waseda University
Identification of daily travel mode preferences under random attention allocation: Evidence from a lab experiment and Paris Abstract Limited and random attention allocation has become an important factor to consider in preference revelation analysis. The Random Attention Model (RAM) relaxes the restrictions of known, accurate attribute information and observable consideration sets, offering potential applications in transportation economics. In this paper, we apply RAM to the transportation setting and provide evidence of internal and external validity. For internal validity, in a lab experiment, we find that the RAM enables us to statistically accept participants’ self-reported preferences and reject “irrational” preferences. The most chosen preference ordering can be explained by the revealed attention intensity at the aggregate level. For external validity, we use GPS- traced daily travel data from 3,319 Greater Paris residents, including mode choice and personal information. We find attention complementary and substitutionary effects across different modes. Private cars receive the most attention, followed by walking, public transportation, private motorbikes, and private bikes. Attention variability is lowest for private cars and walking, while private motorbikes exhibit the highest attention variability. Finally, we find empirically that individuals differ in their most preferred travel modes but show consistency in their least preferred modes.
By WANG Zexuan; Paris Cité Université    Dianzhuo Zhu; University of Lille
Presented by: Dianzhuo Zhu, University of Lille
Behavioral externalities and the hidden cost of fare evasion in Santiago de Chile and New York City Abstract This paper introduces and empirically identifies behavioral externalities, i.e. when other users’ behavior affects utility even when operational attributes remain unchanged. We show the willingness to pay for reducing common public transport misbehaviors (social norm-deviating behaviors) in two cities, focusing primarily on fare evasion. The costs of evasion have been typically analyzed only from a revenue perspective, however, additional (hidden) social costs may arise if it impacts experienced utility, which is not immediately evident, while disutility is more readily expected from other misbehaviors, such as delinquency or informal vending, which have a direct and tangible impact on users. An online discrete choice experiment was conducted in Santiago de Chile and New York City, both of which exhibit particularly high fare evasion rates. We estimate willingness to pay (WTP) to reduce four common misbehaviors: noise from on-board informal musicians, delinquency, informal or unauthorized vending, and fare evasion. The choice experiment was framed as a hypothetical choice between public policies with different expected effects and associated costs. We estimate an Integrated Choice and Latent Variables (ICLV) model that captures: (1) heterogeneous sensitivities to the expected effects of policies on the four misbehaviors; (2) an underlying latent Misbehavior Displeasure that mediates sensitivity to each specific conduct and depends on demographic variables and latent public transport satisfaction; and (3) willingness-to-pay distributions, which allow us to economically evaluate behavioral externalities. We find that delinquency and informal vendors have a generalizable welfare impact, while the impact of fare evasion and informal musicians differs substantially across the two cities. Respondents in Santiago show a positive WTP for reducing evasion, whereas New Yorkers do not exhibit meaningful sensitivity to it. In Santiago, the WTP for reducing the current evasion rate to zero is equivalent to 7.4% of the regular fare, which is nearly half the WTP for eliminating informal vendors or delinquency. Santiago's participants valued a partial reduction of on-board musicians (against a full reduction), while in New York musicians have a positive effect on utility. We discuss the cross-sample differences, the individual characteristics that shape misbehavior disutility, and the policy implications of these results.
By Bastian Henriquez-Jara; Universidad de Chile    Diego Candia; Universidad de Chile    Ricardo Daziano; Cornell University
Presented by: Bastian Henriquez-Jara, Universidad de Chile
Do ride-hailing services compete with or complement mass transit? A multimodal morning commute problem Abstract This study addresses a multimodal morning commuting problem with ride-hailing service consideration, aiming to reveal the relationship between ride-hailing and other travel modes. We begin with a bimodal commuting problem (i.e., departure time and travel mode choices) involving private cars and mass transit. To analyze the impact of ride-hailing on commuting behavior, we extend the model to a multimodal combined-mode-based commuting problem (i.e., departure time and combined mode choices) where ride-hailing is available both as an access mode to mass transit and as a direct mode of travel to the workplace. We examine the equilibrium properties of the model to reveal whether ride-hailing competes with or complements mass transit. We demonstrate that the time cost reduction factor of ride-hailing relative to self-driving and the time-based fare structure of ride-hailing are key determinants of the competitive or complementary relationships.
By Takao Dantsuji; Waseda University    Yuki Takayama; Institute of Science Tokyo, Department of Civil and Environmental Engineer    Zhi-Chun Li; Huazhong University of Science and Technology, China
Presented by: Takao Dantsuji, Waseda University
Session 41: Macroscopic Models of Traffic Flow and Congestion June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 9
Session Chair: Erik Verhoef, VU Amsterdam
Empirical Analysis of Vehicle-Class and Lane-Specific Traffic Flow Dynamics and Economic Pricing: A Case Study of the Fatih Sultan Mehmet Bridge Abstract Traffic flow theory constitutes the backbone of modern transportation economics. The economic analysis of traffic congestion and road pricing has gained substantial importance globally through various congestion pricing implementations. This study investigates macroscopic traffic flow characteristics and their economic implications using high-resolution, event-based data collected from detectors. To capture the true heterogeneity of the traffic flow, the standard vehicle classification utilized by the Türkiye General Directorate of Highways was refined to explicitly differentiate between passenger cars, light commercial vehicles, heavy commercial vehicles, and motorcycles. The preliminary results demonstrate that evaluating the traffic flow as a homogeneous mixed flow is insufficient. Instead, differentiating between vehicle groups and lanes is crucial, as distinct lanes and vehicle types exhibit profoundly different fundamental capacities and economic cost curves.
By Görkem Akyol; MEF University    Goker Aksoy; Gebze Technical University    Onur Tezcan; ITU
Presented by: Görkem Akyol, MEF University
Universal Modeling of Macroscopic Traffic Dynamics in Urban Networks Abstract Macroscopic traffic models are parsimonious tools for large-scale analysis of traffic dynamics in urban networks. While theoretically scalable, the practical application of macroscopic traffic models is often limited by the necessity for model calibration for each network individually. Moreover, the lack of capacity to capture the systemic impact of traffic evolution also limits prediction accuracy for dynamic traffic modeling. To address these limitations, this research leverages machine learning to propose a universal approach for macroscopic modeling of traffic dynamics across diverse urban networks without requiring model calibration for each network individually. By capturing the systemic impact of large-scale network characteristics on traffic dynamics, the proposed deep learning approach enables generalizable modeling of traffic dynamics across diverse urban networks. Our proposed architecture for this machine learning model of macroscopic traffic is a neural network with a static branch for processing, weighting, and encoding network characteristic descriptors, and a sequential branch with an encoder-decoder structure for sequence-to-sequence (seq2seq) estimation of traffic speed dynamics from vehicular density time series at the network level. The proposed model is trained on synthetic data generated by a baseline feedforward neural network model, based on the Macroscopic Fundamental Diagrams estimated across eight major cities in the United States. Following successful training, validation, and testing, the proposed model demonstrates high predictive accuracy and generalizability in modeling traffic dynamics across diverse urban networks. By bypassing the need for network-specific model calibration, the proposed approach facilitates macroscopic traffic modeling, particularly in scenarios where data availability, time, and/or budget are constrained.
By Mahyar Amirgholy; Kennesaw State University    Duleep Rathgamage Don; Kennesaw State University    Sharath Narasimhamurthy; University of Calgary    Kaveh Farokhi Sadabadi; University of Maryland    Lina Kattan; University of Calgary    Jorge Laval; Georgia Tech
Presented by: Mahyar Amirgholy, Kennesaw State University
The Macroscopic Bottleneck Model: An Empirical Exploration Abstract In this paper we provide extensive robust analysis of various alternative specifications of the Macroscopic Bottleneck Model that we initially developed for the Randstad Area in the Netherlands. In particular, besides the initial model which only considered inflows into the network via on-ramps within the area considered for observations in 2019 and 2020, we now consider: •Modelling of evening peak traffic •Specifications that not only consider on-ramps within the area of interest, but also inflows via highways crossing the cordon that defines the area •An extended period of observations, now covering the years 2019-2021 •A revised specification of the congestion function that substantially reduces the multicollinearity problem that plagued the earlier empirical specification
By Erik Verhoef; VU Amsterdam    Bas Turpijn; Rijkswaterstaat    Maarten Verhoef; VU Amsterdam
Presented by: Erik Verhoef, VU Amsterdam
Session 42: Road Pricing in Spatial General Equilibrium June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 11
Session Chair: Diego Candia, Universidad de Chile
A General Equilibrium Model of Zoning, Sorting, and Road Pricing with Agglomeration Abstract Policy debates recognize that congestion, road pricing, mode choice, land use regulation, and agglomeration interact, yet applied models often do not treat these interactions simultaneously. This paper develops a model that integrates zoning, mode choice (including EV), congestion, residential sorting, and agglomeration in a tractable unified general equilibrium framework. The city has a dense Center and a lower density Suburb. Households choose location, housing, goods, discretionary trips, mode (bus, gasoline car, or electric vehicle), and whether to work in the Center or in an outside sector. Preferences are nested CES, yielding price indices and money metric welfare based on realized utilities after sorting. Car travel uses congested roads, and mode choice follows multinomial logit, producing an expected generalized travel cost that enters consumption and commuting. Agglomeration raises the Center wage with aggregate central employment, creating a positive externality, while congestion and housing markets create offsetting forces. Government sets income taxes and transport instruments and returns net revenue lump sum to households. We provide a policy discussion using analytical equilibrium conditions, and quantify full general equilibrium effects with calibrated numerical simulations on welfare, rents, sorting, congestion, mode shares, and VMT. The discussion and simulations provide insights grounded in microeconomics on tradeoffs among policy goals.
By Huibin Chang; British Columbia Government
Presented by: Huibin Chang, British Columbia Government
The welfare effects of road pricing in the presence of non-commuting trips, tax interactions and emissions: a spatial general-equilibrium analysis for the Netherlands Abstract We propose and calibrate a spatial computable general-equilibrium model of the Netherlands to quantify the welfare effects of different road pricing schemes. Households choose a discrete alternative composed of residential and job location, and commuting mode and route. The model also considers the goods production and housing developers sectors. Transport generalized costs feed back into household location, travel, and labor supply decisions, and equilibrium is defined jointly over goods, labor, land and housing markets, and congested travel times. The transport layer distinguises car technologies (ICE, hybrid, electric) and modes such as inter-city rail, local public transport, cycling, and walking. A key methodological contribution is the incorporation of all household-generated traffic, treating non-commuting trips as a direct source of utility. We compute the first-best road-pricing scheme under congestion and emissions externalities, and relate it to Pigouvian, tax interacion and recycling components. We also evaluate the welfare and spatial outcomes under second-best schemes, such as a flat-kilometer charge and some relevant cordon tolls. Preliminary numerical results suggest that explicitly modeling non-commuting trips substantially affects welfare assessments, because mode substitution in non-commuting can be an important margin of behavioral adjustment. Overall, the paper contributes to the literature by evaluating road pricing in a setting with multiple externalities and spatially differentiated marginal utility of income and elasticity of labor supply, by integrating non-commuting travel directly as a source of utility, and by representing local versus highway congestion in a detailed spatial structure.
By Diego Candia; Universidad de Chile    Erik Verhoef; VU Amsterdam    Saúl Basurto; Universidad Nacional Autónoma de México, UNAM
Presented by: Diego Candia, Universidad de Chile
Session 43: Cost-Benefit Analysis of Infrastructure: Biases and Distribution June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 12
Session Chair: Jonas Eliasson, Linköping University
Cost Benefit Analysis is weighted in favor of the wealthiest. What should we do about it? Abstract Socio economic evaluation and its working horse Cost-Benefit Analysis are most often based on the principle of neutrality with regard to distributive issues. This roots in a profound economic reasoning as illustrated in Harberger well-known paper. Yet the ongoing discussion on this topic has left aside a more fundamental consideration: unless specific corrections are implemented in CBA its outcome will not be revenue neutral (giving the same importance to various individuals regardless of their revenue). On the contrary, concepts as willingness to pay and consumer surplus will give more weight to wealthier individuals than to poorer ones. The proposed contribution reviews this issue and proposes theoretically sound solutions. First it defines more clearly concepts related to this issue : revenue neutrality in particular and defines what are redistributive and anti-redistributive outcomes. Second it provides evidences that the routine CBA procedure based on willingness to pay (hence on surplus) gives more weight to wealthier individuals. Third it explores a procedure to correct for the implicit, regressive, welfare weights in CBA. This procedure relies on constant elasticity of substitution utility function. This procedure is fully respecting ordinal utility but provides revenue neutral welfare weights that can be used in evaluation. Next, we apply the method to a prototypical project using real world data.
By Jerome Massiani; Università di Milano-Bicocca    Marco Ponti; Bridges Research Trust
Presented by: Jerome Massiani, Università di Milano-Bicocca
Optimism Bias in Multilateral Development Banks’ Infrastructure Appraisals Abstract The appropriateness of using Cost-Benefit Analysis (CBA) as a decision-making criterion in transport infrastructure planning has been widely questioned due to deviations from appraisal estimates—primarily in investment costs, delivery times, and demand forecasts—and a persistent optimism bias among evaluators that leads to projects with negative social returns, commonly referred to as white elephants. However, there remains relatively little empirical evidence on the extent to which social profitability is affected by these deviations, largely because ex-post evaluations are seldom conducted. In this study, we construct a dataset of infrastructure projects evaluated and financed by Multilateral Development Banks (MDBs) and find that deviations remain modest relative to other studies and evaluators. Most indicators suggest that project profitability is unbiased, that only a relatively small proportion of projects yield negative social returns, and that appraisers from the World Bank show better performance. We argue that this outcome may be attributed to the MDBs’ internal appraisal policies, particularly regarding project governance (with the banks acting as both evaluators and financiers, and applicant countries serving as sponsors) and the mandatory implementation of ex-post evaluations, suggesting that the optimism bias phenomenon is more related to strategic misrepresentation than to a cognitive bias.
By Carlos Parra; Universidad Carlos III de Madrid
Presented by: Carlos Parra, Universidad Carlos III de Madrid
Is the Planning Fallacy the rule or the exception? Frequencies of cost overruns and benefit shortfalls in infrastructure projects Abstract The Planning Fallacy narrative claims that a large majority of projects fail to keep their budgets and deliver their promised benefits. In contrast, the Hiding Hand narrative claims that while projects frequently encounter challenges during execution, they are often overcome and lead to projects that still deliver their promised net benefits. Despite lengthy debates, there is a dearth of empirical evidence regarding which of the narratives best describes reality. We explore this by comparing decision-stage forecasts of both costs and benefits to actual outcomes for all Swedish transport infrastructure investments finalized 1998-2025. In contrast to most of the cost overrun literature, we also explore whether anticipated gross and net benefits were achieved. We find a majority of projects succeed in keeping their budgets and delivering their intended benefits. Most of the failing projects suffer relatively moderate cost overruns. Only a few percent of projects fit the Planning Fallacy narrative, failing both with respect to cost and benefits. Around 15 percent of projects fit the Hiding Hand narrative, going over budget but also delivering more benefits than anticipated, achieving higher net benefits than planned. Most projects fit neither the Planning Fallacy, nor the Hiding Hand; they are, instead, successful.
By Jonas Eliasson; Linköping University    Paul Chapman; Oxford University
Presented by: Jonas Eliasson, Linköping University
Session 44: EVs: Tax Reform and the Fuel-Tax Transition June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 13
Session Chair: Leonardo J. Basso, Universidad de Chile
Electric Vehicle Adoption and the Transition from Fuel Duties to Road Pricing: a Quantitative Spatial Model for Hungary Abstract The transition to electric vehicles (EVs) is reshaping transport systems and public finance. While EVs reduce local environmental externalities, they do not alleviate road congestion, and the spread of them erodes fuel tax revenues, a key source of infrastructure funding. This paper evaluates alternative road pricing instruments in the context of increasing EV penetration using a country-level quantitative spatial model (QSM) for Hungary. We develop a spatial general equilibrium framework featuring (i) households that make nested residential, workplace, vehicle ownership, and commuting mode choices, (ii) monopolistically competitive firms facing iceberg trade costs, and (iii) a construction sector subject to land-use constraints. Key structural parameters are causally estimated using data for Hungary. The fully solved model is used to simulate counterfactual scenarios of EV adoption and road pricing reform. We compare subscription-based access and flat fares with usage-dependent instruments, namely distance-based and congestion-dependent pricing. While non-usage-based schemes may compensate for declining fuel tax revenues, they are unlikely to reduce congestion. In contrast, usage-dependent pricing aligns marginal private and social costs, stabilizes public revenues, and reshapes mobility patterns. Beyond transport demand, the model captures general equilibrium adjustments in wages, land prices, firm entry, and spatial structure. We show how EV adoption combined with well-designed road pricing can improve traffic efficiency, sustain fiscal revenues, and influence long-run urban and regional development in Hungary.
By Réka Kertész-Doffkay; Corvinus University of Budapest    Daniel Hörcher; Imperial College London    Hans Koster; Vrije Universiteit Amsterdam
Presented by: Réka Kertész-Doffkay, Corvinus University of Budapest
Scrappage policies and carbon emission reduction: Evidence from a Swedish car scrappage program Abstract The Swedish car fleet is currently undergoing a major transformation from Internal Combustion-powered vehicles to electric and plug-in hybrid vehicles. To accomplish this, the Swedish Government has implemented two policies: the first is a feebate scheme to increase the sales of new Battery electric (BEV) and plug-in vehicles (PHEVs) and reduce the sale of new ICE vehicles; the second, more recent one, is a scrappage subsidy program that subsidizes scrappage of older ICE cars conditional on purchase of a BEV--used or new--in conjunction with the scrap. Since most cars in circulation in Sweden are not `new', the used car market is an important driver in any greening of the Swedish vehicle fleet, especially with the second policy, scrappage subsidy, explicitly targeting older used cars. Our main objective in this paper is to construct an equilibrium model of new and used cars for Sweden, using administrative data on the universe of car registrations, mandatory vehicle inspection data (which yield mileage data), allied with matched individual and household characteristics of all vehicle owners. We compute the impact of three classes of counterfactual scenarios relating to the two policies, a subsidy unpaired to the purchase of a green vehicle, a subsidy on green vehicles, and the paired subsidy and scrappage scheme. Our estimation, which is ongoing, will use administrative data on the entire universe of registered cars and mandatory vehicle inspection, along with individual and household characteristics of the owners of all private vehicles.
By Chandra Kiran Krishnamurthy; Swedish University of Agricultural Scien    Levi Soborowicz; Swedish University of Agricultural Sciences
Presented by: Chandra Kiran Krishnamurthy, Swedish University of Agricultural Scien
Dynamic Urban Transport Taxes and Subsidies in the Transition to Electric Mobility Abstract The objective of this paper is to analyze dynamic schedules of taxes and/or subsidies to be applied to private vehicles (both combustion and electric) and public transport, with the objective of reducing externalities associated with congestion and transportation emissions, in a context in which EVs are being adopted over time. The policy question then, is what instruments authorities should use over time while on the path to electric mobility. Note that, in addition to varying overtime, taxes could be differentiated or not according to the vehicle’s technology, and subsidies may be considered for EV purchase, and for public transport operation. The primary finding of this study is that, regardless of the scenarios analyzed, a future dominated by EVs is likely to be more congested than the present scenario dominated by combustion vehicles. This outcome is attributed to the low operating costs of EVs. Therefore, proactive evaluation of strategies to mitigate this transportation externality is essential. See full extended abstract attached.
By Francisca Inostroza; Universidad de Chile    Leonardo J. Basso; Universidad de Chile    Maria Börjesson; VTI, LIU    Alejandro Tirachini; Universidad de Chile
Presented by: Leonardo J. Basso, Universidad de Chile
Session 45: Airline Operations, Delays, and Conflict June 19, 2026 9:00 to 10:30 Location: Palazzo Baroni 14
Session Chair: Chiara Morlotti, University of Bergamo
Conflict Effects on African Air Traffic Abstract Air connectivity is vital for economic development in Africa, yet this infrastructure operates in environments frequently disrupted by localized armed conflict. This paper investigates the causal impact of these geopolitical shocks on aviation networks, focusing on both passenger demand and structural centrality. We merge granular flight itinerary data, covering over 16.5 million routes with high-resolution, geocoded conflict data from 2011 to 2024. By exactly reconstructing physical flight edges, we compute monthly weighted centralities and employ a staggered Difference-in-Differences event-study design utilizing a strict zero-casualty control group. Air connectivity is vital for economic development in Africa, yet this infrastructure operates in environments frequently disrupted by localized armed conflict. This paper investigates the causal impact of these geopolitical shocks on aviation networks, focusing on both passenger demand and structural centrality. We merge granular flight itinerary data—covering over 16.5 million routes—with high-resolution, geocoded conflict data from 2011 to 2024. By exactly reconstructing physical flight edges, we compute monthly weighted centralities and employ a staggered Difference-in-Differences event-study design utilizing a strict zero-casualty control group
By Luca Gallarati; University of Bergamo    Laura Ogliari    Davide Scotti; Università di Bergamo
Presented by: Luca Gallarati, University of Bergamo
Flight Delays, Operational Responses, and Their Environmental Cost: Evidence from European Skies Abstract European air transport punctuality worsened markedly in 2023. Beyond passenger disruption and airline operating costs, delays may also generate environmental externalities by increasing realized time and, in turn, fuel burn and CO₂ emissions. This paper quantifies how delays evolve within flights, propagate across aircraft rotations, and translate into additional fuel and CO₂. We use a large flight-segment dataset for Western Europe built from OAG sources (on-time performance, schedules and aircraft information, rotations, and fuel/emissions estimates). We decompose delays into departure, airborne, and arrival components and estimate high-dimensional fixed-effects models to isolate within-flight dynamics and propagation along rotations while controlling for rich route, carrier, and time-varying operational conditions. We document systematic in-flight “recovery”: conditional on a late departure, airborne delay tends to fall, consistent with partial mitigation during the cruise phase. However, propagation along rotations is substantial: arrival delays on one leg significantly increase subsequent departure and arrival delays, indicating cascading effects that amplify disruptions over the day. Finally, longer airborne time is associated with higher fuel consumption and CO₂ emissions per flight, implying a direct environmental channel linking operational disruption to emissions. Overall, mitigation exists but is insufficient to prevent meaningful cascading delays and associated environmental costs. The results highlight that improving operational resilience-reducing reactionary delay and strengthening schedule reliability-can yield both economic benefits and emissions co-benefits.
By Andrea Gualini; Università di Bergamo    Estelle Malavolti; Toulouse School of Economics and ENAC
Presented by: Andrea Gualini, Università di Bergamo
Flight Path Dispersion in Commercial Aviation Abstract This paper investigates dispersion in flight trajectories using global ADS-B data covering flights operated on the first day of each month between 2017 and 2024. We document substantial within-market variation in flown distance, flight time, and speed. While great-circle distance provides a mechanical benchmark for route length, observed operations systematically deviate from this reference due to operational and traffic-related constraints. We examine whether such dispersion is related to congestion, measured as the number of aircraft movements planned within a 15-minute interval at the departure and arrival airports. We estimate linear models relating flown distance, total flight time, average speed, and cruise speed to scheduled congestion, controlling for great-circle distance, aircraft family fixed effects, and origin and destination fixed effects. Results indicate that both departure and arrival congestion are associated with flight operational outcomes. Overall, the findings have important implications for emissions accounting. Specifically, we demonstrate that estimating fuel burn solely on the basis of great-circle distance may lead to systematic underestimation, as actual trajectories are longer and operational adjustments under congestion may further increase fuel consumption.
By Benny Mantin; University of Luxembourg    Chiara Morlotti; University of Bergamo    Renato Redondi; University of Bergamo    Daniel Dobos; University of Luxembourg
Presented by: Chiara Morlotti, University of Bergamo
Session 46: Paper Awards and Celebrating Sergio Jara-Diaz as the 2026 ITEA Honorary Member June 19, 2026 11:00 to 12:00 Location: Palazzo Baroni 16
Session Chair: Jonathan Hall, University of Alabama
Session 47: Autonomous Vehicles: Deployment, Accessibility, and Adoption June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 8
Session Chair: Diana ELHAJJ SLEIMAN, CentraleSupélec - Université Paris Saclay
Autonomous Vehicles and Urban Accessibility: Evidence from Waymo’s Deployment in California Abstract As autonomous vehicle (AV) services expand across U.S. cities, a key open question is whether they improve accessibility for all residents or primarily benefit already well-served communities. This paper provides causal estimates of the impact of AV deployment on residential destination accessibility, exploiting the staggered geographic rollout of Waymo's commercial robotaxi service across California census tracts. Using individual-level mobility data from Veraset combined with deployment records from the California Public Utilities Commission (CPUC), I construct residents' monthly average travel radius as a measure of destination accessibility. I employ a staggered difference-in-differences design following \citet{Callaway2021} to estimate the causal effect of Waymo entry on accessibility. I further examine heterogeneous effects by whether a tract is classified as a disadvantaged community by CalEPA or an equity priority community by MTC, allowing me to assess whether AV deployment narrows or widens existing accessibility disparities. The results speak directly to whether commercial AV deployment, as currently practiced, serves as an equalizing force in urban mobility or reinforces existing transportation inequities.
By Jiatong Xue; University of Alabama
Presented by: Jiatong Xue, University of Alabama
Examining the moderating effects of policy interventions on the adoption of automated public shuttles: evidence from the Norwegian context Abstract Automated vehicles are emerging mobility technologies that have the potential to address the challenges of the current transport sector. Integrating shared automated vehicles (SAVs) into the public transport system is anticipated to provide benefits that support the transition toward more sustainable cities and behaviors. However, social acceptance of these new mobility services remains a key challenge. This study aims to investigate and quantify the moderating effect of three policy measures on behavioral intention to use (BIU) automated public shuttle. Building upon the theoretical frameworks of UTAUT3 and MAVA, we proposed a multi-level research framework to explain behavioral intention to use the automated public shuttle. To validate the proposed research model, we applied the Variance-Based Structural Equation Modeling approach (VB-SEM) using 795 quantitative online survey responses collected in Norway. Our findings reveal that performance expectancy and social influence were positively associated with BIU automated public shuttle. In contrast, perceived risks negatively affect BIU automated public shuttle. Government regulation and user education were found to be essential in facilitating the adoption of these emerging mobility services. However, awareness campaigns have not significantly contributed to promoting the adoption of this new mobility service. Regarding the moderating effects, our results show that government regulation had an insignificant moderating effect on the relationship between perceived risks and BIU. Finally, awareness campaigns and user education contributed significantly to reducing risk perceptions and ultimately enhancing potential users' intention to use. The findings of this study will provide policy implications that could facilitate the adoption of public SAVs.
By Diana ELHAJJ SLEIMAN; CentraleSupélec - Université Paris Saclay    Julie BULTEAU; CEARC,Université de Paris-Saclay, Université de Versailles Saint‐Quentin-en-Yvelines OVSQ    Isabelle Nicolai; Université Paris Saclay, CentraleSupélec
Presented by: Diana ELHAJJ SLEIMAN, CentraleSupélec - Université Paris Saclay
Session 48: Urban Transport Infrastructure and Welfare June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 9
Session Chair: Pierre Picard, University of Luxembourg
Economic Evaluation of a Major Transport Upgrade: Agglomeration, Employment, and Commuting in the SERM Lille Project Abstract In the Hauts-de-France region, commuting between home and work relies heavily on cars, particularly between the mining basin and the Lille metropolitan area. This dependence causes significant congestion on the road network and high greenhouse gas emissions. The Regional Metropolitan Express Service (SERM) project aims to strengthen rail services and intermodality in order to improve accessibility between major urban centers and encourage a modal shift towards public transport. This study examines the extent to which improved rail connectivity associated with the SERM can influence the spatial distribution of employment and places of residence for workers around Lille, as well as the economic gains associated with reduced transportation costs. We use a spatial general equilibrium model derived from quantitative spatial economics. The territory is represented as a set of areas interconnected by transportation costs. Households and businesses choose their location based on accessibility to economic opportunities, transportation costs, and housing market conditions. The SERM is introduced as an exogenous shock reducing generalized rail transport costs. The model parameters are estimated using mobility and socioeconomic data from the Lille metropolitan area, then used to perform a counterfactual analysis. Initial simulations suggest that improving rail connectivity could generate gains in aggregate welfare and lead to a spatial redistribution of economic activity. The results presented here are provided for illustrative purposes. The objective of the study is to calibrate the model in order to estimate the potential magnitude of these effects more accurately.
By Carlos Gbeteglo; Université du Littoral Côte d’Opale, LEM, CNRS UMR 9221, France    Moez Kilani; University of Littoral at Dunkerque
Presented by: Carlos Gbeteglo, Université du Littoral Côte d’Opale, LEM, CNRS UMR 9221, France
Welfare Effects of Congestion in Luxembourg and the Greater Region Abstract This paper studies the effects of congestion relief in a spatial general equilibrium model of Luxembourg and its cross-border commuting zone. Using traffic speed data, we apply a difference-in-differences design on Luxembourg’s highways to measure congestion severity and identify choke points. We then simulate counterfactual scenarios where highway speeds are set to free-flow levels and track the resulting changes in output, welfare, and fiscal revenues. Economic output rises in Luxembourg City and Esch, while other cities lose production but gain in resident welfare. For residents of Luxembourg City, we estimate a short-run welfare loss of EUR 1,140 per person per year, which becomes a welfare gain of EUR 3,490 in the long run after population reallocation. When accounting for migration from the outside economy, the welfare effect in Luxembourg City turns negative at EUR 8,110 per person per year. The elimination of congestion induces a fiscal gain of EUR 2.50 billion per year in the short run, EUR 1.18 billion in the long run, and EUR 7.04 billion when accounting for migration inflows.
By Pierre Picard; University of Luxembourg
Presented by: Pierre Picard, University of Luxembourg
Session 49: Modal Substitution and Multimodal Competition June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 11
Session Chair: Shuo Wang, University of Leeds; Beijing Jiaotong University
Fast Times at Gridlock High: How Does Bundling Services Shift Urban Mobility? Abstract How does bundling services in small locations shift agents’ urban mobility patterns and overall city traffic and welfare? To answer this, I extend a canonical structural quantitative urban model by incorporating 1) non-commuting trips, 2) chained trips, and 3) endogenous traffic costs that depend on all other agents’ movements. I show how to identify and estimate the parameters of this model that govern agents’ daily movements across the city, home and workplace decisions, and equilibrium traffic congestion in the city. I then apply this model to Metro Manila, Philippines, where many varieties of services are bundled together inside malls. Using smartphone data and GPS data that track real-time agents’ movements and traffic in Metro Manila, I estimate the aforementioned parameters, then simulate the creation of new malls in various locations across the province. This allows me to figure out which counterfactual mall locations would minimize traffic congestion and maximize overall welfare in Metro Manila.
By Terry II Culpepper; Harvard University
Presented by: Terry II Culpepper, Harvard University
How Does High-Speed Rail Affect Air Transport over Time? Evidence from China Abstract The rapid expansion of high-speed rail (HSR) has reshaped intercity passenger transport markets and intensified competition with air transport. Since 2008, China’s HSR network has expanded to over 50,000 kilometers, becoming the largest in the world. While existing studies document substantial substitution effects of HSR on air transport—particularly in short- and medium-distance markets—most focus on average or static impacts. Less attention has been paid to how these effects evolve over time. This paper examines the dynamic adjustment of air transport following HSR entry. Using city-pair panel data, a Dynamic Difference-in-Differences (Dynamic DID) framework is employed to estimate event-time impacts on passenger volume, flight frequency, and seat capacity. By allowing treatment effects to vary across relative time, the model traces the temporal evolution of substitution patterns and tests pre-treatment trends. Results are compared with staggered DID and CSDID estimators to assess robustness. The findings indicate a clear substitution effect in short- and medium-distance markets, reflected in sustained declines in air traffic and capacity after HSR entry. The effect strengthens in the early post-entry years before stabilizing. In longer-distance markets, complementary interactions may emerge, and impacts differ across market conditions. The study contributes to understanding the dynamic evolution of modal competition under large-scale HSR expansion.
By Andrew Smith; University of Leeds    Shuo Wang; University of Leeds; Beijing Jiaotong University    Phill Wheat; University of Leeds
Presented by: Shuo Wang, University of Leeds; Beijing Jiaotong University
Session 50: Rural and Urban Mobility: Choices and Consequences June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 12
Session Chair: Patrice Bougette, Université Côte d'Azur, CNRS, GREDEG, France
HOW DOWNTOWN RESIDENTS DELIVER MORE TRANSPORT DAMAGES: SELF-SELECTION AND AVIATION EMISSIONS Abstract As demand for air travel rises and climate change becomes more severe, understanding trade-offs between urban form and travel choices becomes even more relevant for transportation planners and researchers. This paper uses travel data from 129,696 US households (923,572 individual trips) to quantify the carbon emissions and climate impacts of ground-based plus air travel by Americans in 2016 and 2017 as a function of demographics, land use patterns, and other variables. Controlling for self-selection in home neighborhood choices, using a hurdle model for air travel on the NHTS survey day, and recognizing that air travel has additional radiative forcing (from contrails and other atmospheric impacts), results suggest that increasing population density of one’s neighborhood by 1,000 persons per square-mile lowers average (33.3 mi/day) American’s daily driving distance by 0.57 miles(1.71%) and air travel distance by 1.23 miles(4.92%) . At least 34.6% of observed emission differences are due to the household’s location, whereas the remaining 65.4% can be attributed to self-selection. Population-weighted observed averages show Urban residents with the highest mean total CO₂e (63.80 lb), followed by Suburban (58.97 lb), while Rural is lower (47.45 lb). These higher observed emissions in dense areas mainly come from aviation behavior and self-selection, with air travel averaging 45.77 lb per day (1.89% flight rate) in Urban areas versus 13.81 lb (0.57% flight rate) in Rural areas.
By Lewen Bao; The University of Texas at Austin    Kara Kockelman; University of Texas at Austin
Presented by: Lewen Bao, The University of Texas at Austin
What Mobility Solutions for Rural Youth? A Stated Preference Analysis in Low-Density Areas in Franc Abstract This paper explores mobility preferences among young people (15–29) living in low-density areas in France. Using an unlabelled discrete choice experiment on a French representative sample with 1,200 respondents, we assess how they trade off cost, travel time, walking and waiting times, environmental impact, and transport mode - including multimodal and innovative rural solutions. Our modeling exhibits strong preferences for low-carbon options and public transport (especially trains), though private cars remain dominant in rural contexts. Heterogeneity is managed using a conditional logit model with latent classes, allowing the identification of four users profiles with clearly differentiated behaviour. The findings highlight the need for context-sensitive policies: young rural residents respond differently to time and mode constraints than their urban peers.
By Thierry Blayac; Université Montpellier, CEE-M, France    Patrice Bougette; Université Côte d'Azur, CNRS, GREDEG, France    Jérôme Verny; Neoma BS
Presented by: Patrice Bougette, Université Côte d'Azur, CNRS, GREDEG, France
Session 51: EVs: Adoption Incentives and Subsidies June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 13
Session Chair: Yuli Xue, University of Wisconsin - Madison
The Electric Vehicle Transition in Germany: Fiscal Instruments to Offset Public Revenue Shortfalls Abstract As battery electric vehicles gradually replace internal combustion engine vehicles in Germany, public revenues from motorized individual transport - currently approximately 51.2 billion euros per year - will decline substantially. The energy tax on fuels which is currently the largest revenue component, does not apply to electric vehicles, and growing fleet electrification will thus erode this tax base. We examine the fiscal implications of fleet electrification and compare five policy instruments to offset the expected revenue shortfall: a driving electricity tax, road pricing, a vignette, an increased motor vehicle tax, and an increased energy tax on conventional fuels. Using a dynamic simulation model, we derive the required level for each instrument to close the revenue gap. We assess each instrument against five criteria: efficiency, distributional effects, revenue predictability, financing, and transition effects. Road pricing scores highest on efficiency, as it links charges directly to road use. The driving electricity tax is initially progressive, since BEV ownership is currently concentrated among higher-income households. Road pricing and the vignette offer high revenue predictability, whereas the energy tax and driving electricity tax require continuous rate adjustments tied to fleet dynamics. Regarding the transition, only the energy tax increase substantially accelerates electrification - yet it cannot sustain the revenue target beyond 2040, as higher rates erode its own tax base. The paper contributes an integrated comparative assessment of replacement instruments for Germany across policy-relevant dimensions.
By Maria Kennel; University of Münster    Gernot Sieg; University of Münster    Berthold Wigger; Karlsruhe Institute of Technology (KIT)
Presented by: Maria Kennel, University of Münster
How Do Incentive Policies Affect the Adoption of Electric Vehicles in the Ride-hailing Industry? Evidence from Chicago Abstract Transportation Network Company (TNC) drivers face distinct vehicle choice trade-offs relative to private car owners: their high annual mileage—approximately 40,000 miles per year—means that electrifying one TNC vehicle yields roughly three times the emissions reductions of a typical private vehicle. Yet most EV incentive programs treat all buyers identically, and little is known about how high-utilization commercial drivers respond to these broad-based policies. This paper examines how the Illinois EV Rebate Program, which offers a $4,000 rebate for electric vehicle purchases beginning July 2022, affected EV adoption among Chicago ride-hailing drivers. Using monthly administrative data covering the universe of Chicago TNC vehicles from 2015 to 2025, I first document a sharp 0.85 percentage point increase in BEV share at the rebate's introduction—a 62% increase relative to the pre-policy baseline—using a regression discontinuity in time design. To move beyond this reduced-form evidence, I propose to estimate a random-coefficients logit demand model following Berry, Levinsohn, and Pakes (1995) over hundreds of differentiated vehicle models, recovering TNC drivers' preferences over effective purchase price, per-mile operating costs, range, and vehicle characteristics. The structural estimates enable policy counterfactual simulations comparing TNC-targeted rebates, tax credits, and per-mile operating incentives, with cost-effectiveness assessed by comparing program expenditures—including transfers to inframarginal buyers—to the social value of induced emissions reductions. I also evaluate the equity implications of alternative policy designs given the current program's priority for low-income applicants.
By Yuli Xue; University of Wisconsin - Madison
Presented by: Yuli Xue, University of Wisconsin - Madison
Session 52: Market Power in US Airlines June 19, 2026 13:15 to 14:15 Location: Palazzo Baroni 14
Session Chair: Flavio Porta, Università di Bergamo
Untying the Knot: The Competitive Impact of Undoing the Northeast Alliance Abstract This paper analyzes the Northeast Alliance (NEA) − a joint venture between American Airlines and JetBlue Airways at four New York and Boston airports between 2021 and 2023. Under this codesharing agreement, the two airlines jointly scheduled flights and shared revenue. We show that American and JetBlue were significantly more likely to offer codesharing in markets currently served by Delta and United. These findings are a stark contrast to Goetz and Shapiro (2012) who find that codesharing agreements between 1998 and 2010 were a competitive response by legacy carriers to actual and potential competition from low-cost carriers. Using an event study approach, we also document significantly higher prices in NEA markets both during the NEA and surprisingly after the NEA ended. Our findings lend support for the Department of Justice’s decision to challenge this joint venture between American and JetBlue and justification for the District Court’s ruling that the NEA was anti-competitive.
By Nicholas Rupp; East Carolina University    Kerry Tan; Loyola University Maryland
Presented by: Kerry Tan, Loyola University Maryland
Market Power in U.S. Airlines: The Roles of Airport Access and Frequent-Flyer Programs Abstract Airline markets often exhibit prices above marginal cost despite intense competition. This paper quantifies market power and contrasts two mechanisms that may sustain markups: (i) airport capacity constraints (slot control and limited gate access) and (ii) demand-side switching costs related to frequent-flyer loyalty. Using DB1B itinerary-level data for 2022–2024, we estimate a nested-logit demand model and recover product-level markups from Nash–Bertrand first-order conditions. The estimates imply mean own-price elasticities of $-3.20$ and average markups of about \$78. Preliminary results show that terminal gate concentration is strongly associated with higher markups, but this relationship is weaker in slot-controlled markets. Frequent-flyer redemption share is negatively associated with markups on average, with a less negative relationship in slot-controlled airports and systematic differences in tourist destinations. Overall, the evidence points to capacity constraints as the primary correlate of market power, with loyalty effects that depend on market structure rather than uniformly increasing markups.
By Alexander Luttmann; The MITRE Corporation    EVANGELIA PANTELAKI; Emirates Aviation University    Flavio Porta; Università di Bergamo
Presented by: Flavio Porta, Università di Bergamo
Session 53: Pricing and Dynamic Equilibrium June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 8
Session Chair: Alex Anas, State University of New York at Buffalo
Dynamic Pricing for an Autonomous-Vehicle Platoon Lane Abstract This paper develops a modeling framework for the joint management of platoon sizing and toll pricing on a dedicated autonomous-vehicle (AV) lane that also permits human-driven vehicles (HVs) to merge during inter-platoon gaps. Such shared access improves overall lane utilization but introduces a fundamental supply-side trade-off: enlarging platoons raises AV throughput by spreading the fixed inter-platoon safety gap across more vehicles, yet simultaneously reduces the merge windows available to HVs. We derive closed-form expressions for the AV dispatch capacity and the residual HV merge-supply rate as functions of the platoon size and AV entry rate, analytically characterizing this tension through three headway parameters. Traveler behavior is modeled within a dynamic user equilibrium framework in which commuters simultaneously choose departure times and lane assignments, anticipating endogenous congestion on both lanes. The resulting equilibrium is formulated as a mixed complementarity problem with reflected-queue dynamics for both vehicle classes. A tolling extension allows the operator to charge HVs a time-varying toll for platoon-lane access, and the joint platoon-sizing and tolling problem is cast as a Stackelberg bilevel program minimizing total social cost. Preliminary results from a stationary benchmark show that only small platoon sizes (around three vehicles) induce AVs to use the platoon lane, yielding a sharp welfare minimum; larger platoons cause excessive assembly delay, driving AVs away and eliminating platooning benefits. Ongoing work extends the analysis to the full dynamic setting with schedule-delay costs and optimized time-varying toll schedules.
By Xiaoming Li; VU Amsterdam    Erik Verhoef; VU Amsterdam
Presented by: Xiaoming Li, VU Amsterdam
Congestion pricing when distortionary taxes rule Abstract We deal with two questions. 1) How can distortionary taxes be used to imperfectly price congestion? 2) How should road congestion be priced when the distortionary tax rates are preset and cannot be changed? Our simple theoretical setting is that of an urban land economy with two goods, two residence locations and with roads and mass transit in each city, the number of cities endogenously determined. About the first question, we show that a flat rate income tax perfectly prices congestion, while a suburban land tax or a sales tax on the tradeable good do so imperfectly. In all cases, to achieve lower-best efficiency, the full Henry George tax must be used to fund the fixed costs of the infrastructure. About the second question, we show that Pigouvian pricing of congestion can reduce welfare when the preset distortionary taxes have indirectly overpriced the congestion externality. In that case, congestion should be subsidized. We conclude that these insights can be generalized to a spatial general equilibrium setting that can be taken to data.
By Alex Anas; State University of New York at Buffalo
Presented by: Alex Anas, State University of New York at Buffalo
Session 54: Traffic Models, Networks & Simulation June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 9
Session Chair: Mogens Fosgerau, University of Copenhagen
Constructing a population for multi-agent transport simulation: methodological framework and application to the Lille Metropolitan area Abstract The estimation of travel demand is a central task in transportation analysis and planning. In recent years, multi-agent transport simulations have emerged as a powerful framework for representing travel demand as the outcome of individual and household decisions. The reliability of these models, however, critically depends on the quality of the synthetic population used to represent travelers. This paper proposes a methodological framework for constructing a statistically consistent and behaviorally realistic synthetic population for multi-agent transport simulations, with an application to the Lille metropolitan area in northern France. Lille provides a particularly relevant case study due to its dense urban structure, strong commuting flows, and significant cross-border interactions with Belgium, which together generate complex mobility patterns and socio-spatial disparities in accessibility. The proposed framework aims to reconstruct a joint distribution of household and individual characteristics that is consistent with census marginals while preserving the behavioral richness of travel survey data. The approach relies on a household-based reconstruction strategy combining Iterative Proportional Updating (IPU) and weighted resampling. The reconstruction problem is formulated as the minimization of the Kullback–Leibler divergence between the empirical survey distribution and the reconstructed population subject to demographic constraints. Multiple data sources are integrated, including aggregated statistics from census data and detailed behavioral information from household travel surveys. The resulting automated synthesis pipeline produces reproducible synthetic populations while preserving realistic daily activity–travel diaries and intra-household structures. This synthetic population provides a robust empirical basis for calibrating and validating multi-agent transport simulations and for analyzing spatial and socio-economic variations in mobility patterns.
By Moez Kilani; University of Littoral at Dunkerque
Presented by: Moez Kilani, University of Littoral at Dunkerque
Sensitivity analysis of the\\ perturbed utility stochastic traffic equilibrium Abstract This paper develops a sensitivity analysis framework for the perturbed utility route choice (PURC) model and the accompanying stochastic traffic equilibrium model. We derive analytical sensitivity expressions for the Jacobian of the individual optimal PURC flow and equilibrium link flows with respect to link cost parameters under general assumptions. This allows us to determine the marginal change in link flows following a marginal change in link costs across the network. We show how to implement these results while exploiting the sparsity generated by the PURC model. Numerical examples illustrate the use of our method for estimating equilibrium link flows after link cost shifts, identifying critical design parameters, and quantifying uncertainty in performance predictions. Finally, we demonstrate the method in a large-scale example. The findings have implications for network design, pricing strategies, and policy analysis in transportation planning and economics, providing a bridge between theoretical models and real-world applications.
By Mogens Fosgerau; University of Copenhagen
Presented by: Mogens Fosgerau, University of Copenhagen
Session 55: Motorways: Causal Effects & Resilience June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 11
Session Chair: Tomas Paleta, Masaryk University
Motorways and the Farm: Evidence from China’s Motorway Expansion Abstract This paper estimates the causal effect of motorway access on agricultural outcomes in rural China. Using a county-level panel of approximately 2,000 counties across 29 provinces from 2000 to 2020, I exploit the staggered timing of first motorway access as quasi-experimental variation and estimate dynamic treatment effects using the Borusyak, Jaravel, and Spiess (2024) imputation difference-in-differences estimator, which uses 569 never-treated counties as the counterfactual and is robust to cohort-level treatment effect heterogeneity. To circumvent well-documented reporting problems in official rural statistics, I construct a satellite-based measure of agricultural vegetation intensity — the annual mean Enhanced Vegetation Index (EVI) over cropland pixels — from NASA MODIS imagery processed in Google Earth Engine, supplemented by grain output data from the China County Statistical Yearbook. The results reveal a striking divergence: motorway access raises EVI by approximately 2.6% of the sample mean, with positive effects emerging immediately upon treatment and persisting through six post-treatment years, while simultaneously reducing grain output by up to 0.15 log points. This divergence between rising vegetation intensity and falling grain production is inconsistent with a simple aggregate productivity gain and instead points to compositional change within agriculture — a motorway-induced reallocation toward higher-value, transport-sensitive crops with greater biomass, as labor-intensive grain cultivation contracts. These findings suggest that China's expressway expansion reorganized the structure of farming rather than uniformly raising agricultural productivity.
By Ting Liu
Presented by: Ting Liu,
Cross-Border Resilience of Central European Motorway Corridors Under Compound Shocks: COVID-19 and the Outbreak of the War in Ukraine (Evidence from Czech and Slovak Toll Systems) Abstract Central European motorways connecting Czechia and Slovakia, serve as critical arteries of the TEN-T core network, handling both freight and long distance passenger flows. This paper examines their resilience under two consecutive macro-shocks: the COVID-19 pandemic (2020 lockdowns) and Russia's invasion of Ukraine (February 2022). Using a harmonized dataset of toll system transactions from 2019–2022, we analyze high-frequency traffic volumes for heavy goods vehicles (trucks) and passenger buses/coaches across border gateways, interior segments, and key corridors. We implement a dual-event study design, estimating dynamic responses to the March 2020 lockdown onset and the February 24, 2022 invasion. Difference-in-differences comparisons isolate border effects, while spatial analyses test for network re-routing. COVID-19 produced asymmetric shocks: deep, persistent bus traffic collapses versus moderate truck discontinuities from cross-border frictions. The war shock inverted this pattern, generating unprecedented eastward-to-westward bus surges from refugee evacuations and structural truck reallocation as eastern trade routes collapsed under sanctions and energy shocks. Preliminary findings reveal an East-West gradient in 2022: eastern gateways saw acute capacity strain, with substitution onto parallel North-South corridors. Corridors with high pre-shock redundancy exhibited faster recovery. Truck resilience during COVID highlights supply chain robustness, while bus fragility underscores passenger services' dual role in routine connectivity and crisis mobility. This granular evidence informs TEN-T policy, advocating dynamic border management, corridor redundancy investments, and recognition of coach services' emergency utility. By contrasting demand-collapse and demand-surge shocks within an interconnected network, we identify actionable vulnerabilities for future compound crises.
By Tomas Paleta; Masaryk University
Presented by: Tomas Paleta, Masaryk University
Session 56: Transport Affordability June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 12
Session Chair: Francesca Giacco, DLR (German Aerospace Center)
SOCIO-ECONOMIC STRESS IN RELATION TO CAR TRAVEL NEEDS AND TRANSPORT ALTERNATIVES: WHAT IS THE CONTRIBUTION OF ENERGY COSTS? Abstract Fuel poverty and vulnerability have emerged as distinct concepts, respectively to characterize current household’s inability to achieve their energy needs at reasonable cost with respect to their resources, and socio-economic risks associated with rising fuel prices. In the field of transport, they can manifest under the forms of either budgetary stress or travel restrictions, none of which is directly observable, but can be revealed through the estimation of car travel needs, while also accounting for transport alternatives. In this research, we resort to a Tobit model of daily car travel needs to reproduce the annual required car mileage distribution among French households through a Monte-Carlo simulation, which allows us to estimate the frequency of travel restrictions, along with degrees of related budgetary stress among households. Our results point out the contribution of rising car travel needs to the increasing frequency of travel restrictions, although public transportation alternatives allow to mitigate their severeness, especially in large cities. The variable adequacy of alternative transport supply across residential areas tends to reinforce the advantage of high-income groups through their ability to locate in central, well-desserved areas, thus contributing to reduce their vulnerability. By contrast, situations of transport-related socio-economic stress are found mostly among the poor and the lower middle class. The contribution of energy costs to transport-related socio-economic stress is modest in general, but more significant among low-income households and residents of low-density areas, in particular in small to medium urban areas and the rural space.
By Richard GRIMAL; CEREMA
Presented by: Richard GRIMAL, CEREMA
Estimating transport expenditure shares to assess households' risk of being unable to afford transport Abstract Transport affordability has become an important concern of policy makers as many households are affected by increasing energy prices in the last few years. A commonly used indicator of transport affordability in the literature is whether a household’s share of expenditure for transport is below 6% of its total expenditure. We estimate a hierarchical Bayesian Beta regression model to estimate transport expenditure shares for households in different circumstances using a sample of ca. 40,000 households. This approach allows quantifying the uncertainty of a household's group average expenditure being above the threshold. The results show that the main factor influencing crossing the 6% threshold is car ownership, especially for the lowest income groups. Other factors as having children and being employed also contribute to heterogeneity in expenditure shares, while regional differences alone do not seem to play a big role. This study shows how providing a disaggregated picture of which households and with which probability they are at risk of being unable to afford transport can help policy makers better informed decisions.
By Francesca Giacco; DLR (German Aerospace Center)    Ariane Kehlbacher; DLR - German Aerospace Center
Presented by: Francesca Giacco, DLR (German Aerospace Center)
Session 57: EVs: Consumer Behavior and Use June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 13
Session Chair: Andrea Pellegrini, Institute of Transport and Logistics Studies
Losing the Fast Lane: Travel Behavior Impacts of Removing EV Access to Public Transport Lanes in Oslo Abstract Norway has developed a highly mature Battery Electric Vehicle (BEV) market through extensive incentives, including free access to public transport lanes (PTLs). A recent policy change banned BEVs from utilizing PTLs in Oslo and Akershus counties, providing a unique opportunity to study driver reactions to the withdrawal of such privileges. This study investigates the impact of removing this benefit on driving behavior and modal split. Utilizing high-resolution GPS tracking data and socio-demographic surveys, we employ a Difference-in-Differences (DiD) approach to compare drivers with high versus low pre-policy reliance on PTL-equipped roads. Early results reveal highly exposed drivers reduced travel on these roads by at least 20%, suggesting a rerouting effect. For trips contained entirely within the affected counties, there is a change in the modal split, marked by an increase in public transport and a reduction in car use, primarily during rush hours. Conversely, when including longer trips, while public transport usage also increases, total car distances potentially grew as drivers adapted through alternative, longer routes and off-peak rescheduling. Ultimately, these findings provide empirical evidence on how eliminating established BEV privileges influences route choice and modal split within a mature market.
By Giacomo Seravalli; University of Siena    Alice Ciccone; Institute of Transport Economics    Christian Weber; Institute of Transport Economics
Presented by: Giacomo Seravalli, University of Siena
FROM AWARENESS TO PARTICIPATION: THE ROLE OF CONSUMER UNDERSTANDING IN VEHICLE-TO-GRID BENEFIT EXPECTATIONS Abstract Vehicle-to-grid (V2G) has been widely promoted as a mechanism for integrating electric vehicles (EVs) into electricity markets by enabling bidirectional energy flows. While much of the existing literature focuses on technical feasibility and system-level value, comparatively less attention has been paid to how EV owners perceive the private financial benefits associated with participation. This study examines how awareness and conceptual understanding of V2G shape expected annual electricity bill savings among EV owners. Using data from a nationwide survey of Australian 1,794 EV owners, we distinguish between awareness of V2G and correct conceptualisation of the technology. We estimate a sample-selection endogenous ordered probit model that jointly accounts for non-random awareness and endogeneity in correct understanding when modelling perceived savings. Results indicate that awareness and understanding are distinct phenomena and that correct conceptualisation is strongly associated with higher expected financial benefits. Predicted mean perceived annual savings approximately double when respondents correctly define V2G. At the same time, overall expectations remain modest, with most EV owners anticipating incremental rather than substantial bill reductions. Significant heterogeneity emerges across segments, particularly by solar panel ownership, battery storage ownership, home charging access, and age. The findings suggest that informational frictions and incomplete understanding may act as behavioural constraints on V2G participation, even in a context characterised by high residential solar uptake and widespread home charging. From a policy perspective, the results underscore the importance of clear communication, contract transparency, and targeted program design in scaling V2G participation.
By Andrea Pellegrini; Institute of Transport and Logistics Studies    Elisabetta Cherchi; NYU Abu Dhabi    John Rose; The University of Sydney
Presented by: Andrea Pellegrini, Institute of Transport and Logistics Studies
Session 58: Cabotage and Dynamic Capacity June 19, 2026 14:30 to 15:30 Location: Palazzo Baroni 14
Session Chair: Laingo Randrianarisoa, Kedge Business School
The International JetBlue Effect: Consumer Gains from Entry in Transatlantic Air Travel Markets Abstract JetBlue entered transatlantic air travel markets at comparable prices to competitors, which resulted in an increase in the total number of travelers and a minimal reallocation of existing shares—a phenomenon we dub the “international JetBlue effect.” We estimate that JetBlue’s entry into 13 transatlantic markets from its New York and Boston hubs has generated roughly $259 million per year in consumer surplus, with approximately 38% of the benefits accruing to the business class cabin. We simulate JetBlue’s entry into 16 similar transatlantic markets and estimate that such entry would generate an additional $108 million per year in consumer surplus. We suspect that the international JetBlue effect could arise as other US carriers expand into international markets, such as Alaska Airlines’ planned expansion into Asia from its US West Coast hubs.
By Austin Drukker; Federal Trade Commission    Kerry Tan; Loyola University Maryland    Nicholas Rupp; East Carolina University
Presented by: Austin Drukker, Federal Trade Commission
Dynamic Capacity Adjustments: The Case of the U.S. Domestic Airline Market Abstract We examine dynamic capacity adjustments in U.S. domestic airline markets over the period 2010-2020. Capacity decisions are a central element of airline strategy. They can be adjusted along several aspects such as flight frequency, aircraft size, and seating density, each with different implications for costs, competition, and service quality. This research investigates how market performance, competitive conditions, and cost factors relate to subsequent capacity decisions. In our empirical analysis, we carry out fixed-effects regressions with lagged explanatory variables to capture the dynamic relationship between observed market outcomes and capacity adjustments. An airline’s capacity in a market is measured using available seat kilometers (ASK), while market performance and market structure are characterized by using passenger volumes (both non-stop and one-stop connecting passengers), revenue per available seat kilometer (RASK), concentration measures, low-cost carrier presence, hub indicators, and cost variables. We document systematic patterns in how airlines adjust capacity across markets and over time, contributing to the empirical literature on airline competition and capacity planning.
By Laingo Randrianarisoa; Kedge Business School    Benny Mantin; University of Luxembourg
Presented by: Laingo Randrianarisoa, Kedge Business School
Session 59: Structural Spatial Models June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 8
Session Chair: Minoru Osawa, Kyoto University
The gravity of urban leisure destinations: empirical results from a quantitative spatial model Abstract Non-commuting trips represent the majority of trips in developed cities. To assess their importance, we develop a quantitative spatial equilibrium model that features non-commuting trips to non-work activities. The theoretical model predicts a gravity structural equation for non-commuting trips that depends on how attractive and accessible a destination is relative to any other destination in the city. Additionally, the theoretical model also predicts that the accessibility to non-work activities matters for the residential and workplace choices of residents in a city. Using large-scale GPS data for the city of Budapest, we estimate the key parameters governing the gravity equation for non-commuting trips. This allows us to assess its importance at determining the residential and labor choices of residents relative to more canonical variables such as amenities, rents, wages or commuting costs.
By Daniel Ruiz Palomo; Universitat Autònoma de Barcelona    Daniel Hörcher; Imperial College London
Presented by: Daniel Ruiz Palomo, Universitat Autònoma de Barcelona
Transport Costs and the Dynamic Stability of Regional Agglomeration Abstract Transport improvements can either concentrate economic activity in major hubs or spread it across regions, but their long-run effects are not well understood. This paper develops a dynamic multi-region model with migration and intergenerational effects to study how transport costs influence the stability of regional structures. Starting from a stable dispersed equilibrium, lower transport or migration costs can destabilize the system and lead to different adjustment paths. The results show that stable outcomes usually involve concentration in a single core region, while polycentric patterns tend to appear only as temporary or oscillatory dynamics. Overall, adding dynamics does not fundamentally change the tendency toward concentration found in static models. These findings suggest that transport investments may reinforce regional inequality unless additional stabilizing forces are present.
By Minoru Osawa; Kyoto University
Presented by: Minoru Osawa, Kyoto University
Session 60: Transit Service Design and Competition June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 9
Session Chair: Alejandro Tirachini,
Non-price Competition in Deregulated Urban Bus Systems Abstract This article attempts to explain why the frequency is high and the capacity of buses goes down when the urban public transport market is not regulated. To do so, it is assumed that bus services compete on frequency and capacity, given a flat rate. The paper compares the effect of different regulation schemes on the provided frequency and the consumer surplus. This comparison allows us to identify tools for regulating urban bus services. The proposed model assumes that bus lines simultaneously serve two markets: one in which they are monopolistic and one in which they compete for demand. Equilibrium frequencies increase with total demand in the competitive segment and marginal net revenue per passenger. Moreover, the rents obtained from the competitive segment should dissipate because the game in frequency is a contest. We show that bus capacity decreases with total demand in the competitive segment and net revenue per passenger. Also, firms have incentives to reduce bus capacity when they are not dominant and to increase it when they are. Therefore, deregulation leads the non-dominant firms to reduce the bus size in the long term.
By Marco Batarce; Universidad San Sebastian, Chile
Presented by: Marco Batarce, Universidad San Sebastian, Chile
Optimal design of public transport services and allocation of street space considering the health benefits from walking and cycling Abstract The objective of this research is to evaluate the benefit of implementing subsidies for active mobility and quantify the effect of accounting for health benefits in the determination of optimal supply levels and pricing for public transport and optimal allocation of street space between cars, buses and bicycles. In particular, we study if monetised health benefits have an influence on the desirability of cycleways and busways in situations with limited street space available, using a social welfare maximisation model. Optimization variables are the bus fare, frequency, bus size, car pricing, and subsidies for active mobility. We formulate first-best and second-best scenarios. The application of the model to a large avenue with four available lanes per direction in Santiago shows that the consideration of health benefits is critical to the resulting desirability of having a full lane dedicated to bicycles. Different scenarios with segregated busways are also examined, together with the effects on the optimal frequency of service in public transport. We find that optimal public transport fares indeed depend on the existence of health benefits in active mobility through two mechanisms: (i) the health benefits from walking as an access mode tend to reduce the bus fare, but (ii) the health benefits from walking and cycling as alternative modes to public transport tend to increase the bus fare in second-best, when cycling and walking are not subsidised. Implications for policy making, in particular regarding policies to subsidise public transport and active mobility, are discussed.
By Benjamin Sabag; Universidad de Chile    Alejandro Tirachini    Ricardo Hurtubia; Universidad Catolica
Presented by: Alejandro Tirachini,
Session 61: European Rail: Regional and Cross-Border June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 11
Session Chair: Jakub Daněk, Masaryk University / Université Libre de Bruxelles (ULB)
A Direct Ridership Model for regional railways in Lombardy Abstract This paper aims to assess the factors affecting ridership of regional railways in Lombardy, Italy. This is done building a Direct Ridership Model (DRM) analysing the demand at the station level as a function of a variety of internal and external factors. To do so, we develop a regression model starting from morning peak ridership data for an average workday in May 2024. In this contribution, we focus on a couple of innovative features included in the model. First, our work concerns a regional railway system crossing a variety of territorial contexts, instead of more commonly studied urban and metropolitan networks. Second, we use network centrality indicators to give a more precise description of the supply, accounting for the possible destinations to be reached as well as the travel time needed to reach them. The model developed has a double potential utility. First, to give empirical insights for policymakers to address a variety of issues, for example related to how people access the stations. Second, to have a simple but powerful instrument for simulating demand variations as a response to changes in internal or external conditions.
By Tommaso Battilocchi; Politecnico di Milano    Paolo Beria; Politecnico di Milano    Lorenzo Mattioli
Presented by: Tommaso Battilocchi, Politecnico di Milano
Border effects in Central European rail passenger transport Abstract Within the European Union, the free movement of people and goods is formally guaranteed, and there are no institutional barriers to cross-border transport. Therefore, at least in theory, border crossings should not represent a significant obstacle to transport flows. However, empirical evidence from cross-border traffic movements suggests otherwise. Despite the absence of formal restrictions, national borders continue to constitute a substantial barrier to the movement of both people and goods. The negative impact of borders is frequently examined in trade literature. In the field of passenger transport, there are only a few studies, most of which concern air transport. The impact of borders on land transport is practically unexplored. To fill this gap, the aim of this study is to provide estimates of border effects in rail transport. An extended gravity model is estimated using detailed data on the daily frequency of rail connections on 48 routes in Central European countries, observed over 7 consecutive days. Control variables include typical gravity variables such as population, GDP, and distance, as well as other factors such as connection speed and intermodal and intramodal competition. Preliminary results show that borders reduce frequency by about 66%. In addition, it was found that the magnitude of border effects varies across country pairs.
By Jakub Daněk; Masaryk University / Université Libre de Bruxelles (ULB)    Zdeněk Tomeš; Masaryk University    Vilém Pařil; Masaryk University
Presented by: Jakub Daněk, Masaryk University / Université Libre de Bruxelles (ULB)
Session 62: Fuel and Energy Pricing: Welfare Effects June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 12
Session Chair: Misak Avetisyan, Texas Tech University
Do resident discounts benefit consumers? Unintended consequences of cross-border energy policies Abstract This study examines the impact of resident fuel discounts on gasoline prices and market competition in cross-border regions. Using a spatial competition model, we explore how these discounts influence fuel station pricing strategies and consumer be- havior. Although resident discounts may initially benefit consumers, they can also lead to unintended price increases and consumer harm by reducing competitive pres- sure among local fuel stations. To validate these insights, we analyze gasoline prices from the Friuli-Venezia Giulia region in Italy, where resident fuel discounts aim to coun- teract fuel tourism. Employing a fixed-effects regression approach, we find a significant positive relationship between discount levels and gasoline prices, suggesting that fuel stations adjust prices upward in response to the policy. Price sensitivity varies by proximity to the Slovenian border, with stations closer to the border exhibiting greater responsiveness. Our findings highlight the complexities of fuel pricing policies and their unintended consequences, and stress the importance for policymakers to take into ac- count the possible reduction in competition that can stem from residential discounts.
By Angela Stefania Bergantino; Università di Bari    Mario Intini; Università di Bari    Flavio Pino; Politecnico di Torino
Presented by: Mario Intini, Università di Bari
Impacts of Fuel Taxation on Modal Choice, Emissions and Welfare Abstract In this study we explore the potential impacts of fuel taxation on transportation emissions within the major metropolitan cities in the United States. The application of fuel taxation is modeled as an exogenous economic shock that stimulates changes in the consumer and producer behavior. We focus on shifting to other transportation means, economic implications, and environmental gains. The results indicate that one unit increase in the U.S. gasoline price results in significant reduction in U.S. and global transport emissions by 0.66% and 0.12%, respectively, especially when shifts to less emission intensive transport modes are considered. Our findings show that fuel taxation is an efficient means aimed at mitigating U.S. and world greenhouse gas emissions by 0.16% and 0.02%, respectively. We also find that other countries exhibit disproportionate welfare effects with increased fuel taxes in the United States. This study provides policy makers evidence for employing market-based instruments to improve the environmental quality and it can be an important steppingstone for sustainable transportation planning.
By Misak Avetisyan; Texas Tech University    Ali Jaffri; College of Business, North Dakota State University
Presented by: Misak Avetisyan, Texas Tech University
Session 63: EVs: Charging Markets and Infrastructure June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 13
Session Chair: Jingjing Li, Vrije University Amsterdam
Charging interoperability in the electric vehicles industry: vertical integration and market foreclosure Abstract This paper investigates the regulation of public charging infrastructure (CI) and charging services for electric vehicles (EVs) to enhance consumer surplus and social welfare. We analyze the strategic interaction between Charging Point Operators (CPOs), who invest in CI, and Mobility Service Providers (MSPs), who deliver charging services to end users, considering both vertically integrated and separated market structures. Our model incorporates indirect network externalities: an expanded CI network reduces range anxiety, thus increasing EV adoption, while rising EV demand enhances the profitability of CI investments. The analysis highlights a fundamental trade-off between allocative efficiency and market competition. Vertically integrated CPO-MSP firms tend to engage in foreclosure, limiting access to independent MSPs and reducing downstream competition. Despite this anti-competitive behavior, vertical integration can lead to efficiency gains due to internalization of the complementary investments, resulting in greater CI network coverage and higher EV demand relative to separated markets. We further explore the impact of two distinct government subsidy schemes: consumer-targeted EV subsidies (CTS) and firm-targeted CI subsidies (FTS). The relative efficacy of these policies in stimulating CI investment, expanding EV adoption, and enhancing consumer surplus depends critically on the degree of vertical integration. Specifically, CTS policies are more effective in integrated markets, while neither subsidy scheme fully addresses the trade-off between increased social welfare and diminished competitive intensity.
By Tiziana D'Alfonso; Sapienza University of Rome    Mirko Giagnorio; Sapienza University of Rome    Maria Börjesson; VTI, LIU
Presented by: Mirko Giagnorio, Sapienza University of Rome
Travel-Demand-Based Tradable Capacity Permits for Managing EV Charging Congestion Abstract The rapid diffusion of electric vehicles (EVs) is transforming electricity demand and creating new challenges for electricity distribution networks. Residential charging is often highly concentrated during evening peak hours when commuters return home, generating substantial stress on local distribution infrastructure. At the same time, EV users typically have access to multiple charging options, such as home and workplace charging, which differ in convenience and timing and are often provided by competing operators. This paper develops a stylized economic model of EV charging to analyze how charging demand interacts with distribution network capacity constraints and market competition between charging providers. In the model, electricity is produced upstream and delivered through a regulated electricity utility to two downstream charging operators that compete in prices for EV users. Users choose charging location, time of day, and quantity to maximize utility, while electricity supplied at different locations and times is modeled as imperfect substitutes. The analysis characterizes decentralized equilibrium charging patterns and examines the welfare implications of alternative peak-demand management instruments. In particular, we compare time-of-use pricing with tradable capacity permits that allocate scarce peak-period distribution capacity through a market mechanism. The model is further extended to consider vehicle-to-grid (V2G) participation, which allows EV users to discharge electricity during peak periods and provide additional flexibility for congestion management. The results provide insights into how market structure and policy design interact in managing EV charging demand under distribution network constraints.
By Jingjing Li; Vrije University Amsterdam    Erik Verhoef; VU Amsterdam    Yacan Wang; Beijing Jiaotong University
Presented by: Jingjing Li, Vrije University Amsterdam
Session 64: Airport Performance and Gates June 19, 2026 16:00 to 17:00 Location: Palazzo Baroni 14
Session Chair: Viviane Falcao, Federal University of Pernambuco
Gate Access, the Hub Premium, and Annual Gate Reallocation: Evidence from Seattle and San Francisco Abstract Using reduced-form and structural econometric approaches, this report examines how airport barriers such as access to gates at the origin and destination airports impact the fare premium that airlines charge passengers at their hub airports. Combining transacted fare data from the Department of Transportation with information on gate access from OAG Flightview across pre-covid (2016Q1–2019Q4) and post-covid (2022Q1–2023Q4) periods, we find that a substantial fraction of this “hub premium” is due to product differentiation (e.g., flight frequency) and access to airport gates. More importantly, we find that fare increases that occur in conjunction with increased gate access are mitigated at Seattle-Tacoma International and San Francisco International, two airports that reallocate common use and preferential use gates annually. This finding implies that annual reallocation of gates reduces barriers to entry, benefiting consumers via lower fares. Accordingly, annual reallocation of gates is a policy that should be considered by airport authorities who wish to decrease fares for passengers flying to and from their airports.
By Alexander Luttmann; The MITRE Corporation
Presented by: Alexander Luttmann, The MITRE Corporation
RESEARCH PROTOCOL FOR AIRPORT EFFICIENCY BENCHMARKING ANALYSIS Abstract This study addresses critical limitations in current airport benchmarking methodologies, specifically financial data opacity, static bias inherent in annual models, and structural heterogeneity among airport clusters. Through a systematic literature review following PRISMA 2020 guidelines, we analyzed 68 high-impact studies published between 2020 and 2026 to propose a comprehensive research protocol integrating technical efficiency, financial performance, and Environmental, Social, and Governance (ESG) dimensions. Our findings reveal a clear methodological evolution from static Data Envelopment Analysis (DEA) models to hybrid architectures incorporating Machine Learning for predictive analysis and ESG-integrated models measuring environmental sustainability. This protocol provides a transparent, data-driven framework for regulatory monitoring and investment decision-making in the aviation sector, offering practical implications for policymakers, airport operators, and investors seeking to enhance operational efficiency while promoting sustainable development.
By Viviane Falcao; Federal University of Pernambuco    Hélio da Silva Queiroz Júnior; UFBA    Gustavo Barbosa; UFPE    Leonardo Silva; Federal University of Pernambuco    Juliana Lima; Universidade Federal do Ceará    Michelle Bandeira; Universidade Federal de Goias    Francisco Silva; Universidade Federal do Ceará
Presented by: Viviane Falcao, Federal University of Pernambuco

 

This program was last updated on 2026-06-03 17:59:28 EDT