When 48 nations descend on six North American cities over six weeks, their transit systems face the most concentrated demand spike of the decade. This report applies causal inference methods to measure ridership impact, map fare equity, and model system strain across Toronto, Vancouver, New York/New Jersey, Atlanta, Seattle, and Mexico City. It is a living document, updated as data arrives.
At 1pm local time on June 11, Mexico City hosted the tournament's opening match at Estadio Azteca. Eighty-seven thousand fans packed the oldest World Cup venue in history. Roughly 50,000 of them got there on the Mexico City Metro, at $0.28 per ride.
Five time zones east, in New Jersey, tickets to the July 19 Final at MetLife Stadium were already reselling above $3,000. Getting to MetLife requires a different calculation. NJ Transit announced a special World Cup train-and-shuttle package at approximately $150 per person from Secaucus Junction. The standard roundtrip runs about $25. At New York City's minimum wage of $17/hour, that $150 package costs 8.8 hours of labour to afford. The Metro ride in Mexico City costs 20 minutes of the local minimum wage.
The real price of the World Cup is not in the ticket. It is in the corridor between your door and the stadium gate. Paraphrasing transit equity research on mega-event fare design
This report applies econometric methods from public-finance and urban economics to document, estimate, and explain that gap, city by city, match by match, dollar by dollar. The core question is not whether transit gets stressed during a World Cup. Of course it does. The question is how much, where, and what choices each city made about who bears the cost.
Three questions drive this analysis. First: does hosting World Cup matches increase transit ridership, and by how much? Second: is that increase large enough to strain corridor capacity? Third: what does it actually cost to get to the stadium, and who pays the highest burden?
Each question requires a different method, different data, and different caveats. Here is a plain-language summary of what we do and why.
Why three methods and not one? They answer different questions. The event study tells us what happens station by station, hour by hour. The panel DiD tells us whether the host city's system as a whole sees a ridership lift. The synthetic control gives the cleanest causal estimate of the total effect. Each method has blind spots the others cover.
FIFA 2026 spreads 104 matches across 16 North American cities. This analysis focuses on the six where granular, publicly available transit data enables causal identification and where the range of system types, densities, and fare structures creates the sharpest contrasts.
MetLife has no direct rail connection to New York City. Fans from Manhattan take NJ Transit from Penn Station to Secaucus Junction, then a dedicated shuttle to the stadium. This is the only option. There is no subway, no direct commuter rail, no alternative rail route. The system was never designed for simultaneous sell-out events at one venue on that corridor.
The $150 Secaucus VIP package bundles guaranteed boarding, dedicated platform access, and shuttle service. Standard NJT service runs at normal fares. Critics argue this creates a two-tier transit experience priced explicitly to extract surplus from captive demand. Defenders point out that NJT is not removing standard service; it is adding a premium tier. The underlying constraint is physical: limited track capacity means someone controls access via mechanism, and price is the mechanism NJT chose.
The Final on July 19 is the single highest-strain event in this dataset. 82,500 fans, 45% arriving by transit, exiting a stadium served by one rail corridor. No redundancy. No parallel route. Verify current NJT pricing at njtransit.com before citing specific figures.
The Mexico City Metro's Line 2 terminates at Estadio Azteca station, one of the most direct stadium-transit connections in the world. A single ride costs $0.28 USD (roughly five Mexican pesos). For the opener between Mexico and South Africa, approximately 49,000 fans are projected to arrive via Metro. Line 2 runs at roughly 40 trains per hour during event service, which is why strain here sits below Vancouver despite a much larger stadium.
The fare equity picture in Mexico City inverts every other city in this study. Getting to the Azteca costs less in absolute terms than a single station transfer on the New York subway. That says as much about New York as it does about Mexico City. The CDMX Metro is a heavily subsidised public good; its fares have not kept pace with operating costs for decades. That subsidy makes the World Cup remarkably accessible for local fans. International visitors paying $0.55 for a stadium roundtrip are benefiting from a political choice about who should fund urban mobility.
BMO Field at Exhibition Place is served by TTC routes 509 and 511 streetcar, with GO Transit connections from Union Station. At roughly 45,700 seats, it is the smallest World Cup venue in this sample. Toronto's high transit mode share and compact urban form keep the strain index below other cities despite solid ridership. The stadium simply cannot fill as many transit riders as MetLife or the Azteca can.
TTC projects 3.8 million total boardings attributable to World Cup events, the highest headline figure of any Canadian city. For context, a typical weekday across the full TTC network is around 1.1 million trips, so these additional match-day boardings represent a meaningful seasonal boost spread across six weeks. This is Toronto's largest sports-transit mobilisation since the 2015 Pan Am Games.
Vancouver has the strongest stadium-transit integration in North America for this tournament. BC Place is a two-minute walk from Stadium-Chinatown station, served by both the Expo and Millennium SkyTrain lines. TransLink's decision to include transit in FIFA match tickets mirrors the Qatar 2022 free-transit policy and Vancouver's own playbook from the 2010 Winter Olympics.
The $21.6M CAD investment funds roughly 600 additional bus trips per day and enhanced SkyTrain frequency across the tournament. Even so, the 80% mode share and the physical limits of SkyTrain headways post-match push the strain index above 7. Corridor crowding is a real concern even with the best planning in North America for this event. The question is whether crush loading on SkyTrain is preferable to gridlock on downtown Vancouver roads. Based on prior events at BC Place, the answer is yes, though egress times still run 45 to 60 minutes.
Atlanta is the most car-dependent city in this sample. Mercedes-Benz Stadium sits adjacent to GWCC/CNN Center station on MARTA's Red and Gold lines, which is actually better connectivity than many Atlanta destinations. But a typical event mode share of under 30% reflects decades of car-oriented urban form. International visitors unfamiliar with Atlanta's sprawl will likely shift that number upward during the World Cup, making MARTA's capacity and fare structure unusually visible.
Georgia's minimum wage sits at the federal floor of $7.25/hour, unchanged since 2009. A $5 MARTA roundtrip is 69% of one hour's minimum wage. That figure is the highest affordability burden among the US cities in this dataset, not because MARTA charges unusual fares (it doesn't), but because the denominator has not moved in 17 years. Atlanta hosts a Semi-final on July 15 in addition to five group stage games, making it, alongside MetLife, the busiest transit test in this analysis.
Lumen Field is served by International District/Chinatown station on the Link Light Rail, connecting to Sea-Tac Airport, Capitol Hill, University District, and Northgate. Geographic coverage is relatively broad for a system still expanding. King County Metro buses add capacity from West Seattle, the Eastside, and South King County.
The post-match dispersal challenge is acute. Seventy-two thousand fans must leave a stadium on a narrow waterfront strip, with a single rail corridor as the primary alternative to gridlocked surface streets. Sound Transit is running enhanced Link frequency on match days, but the 30-minute egress window is tight. Seattle's group stage includes USA vs Australia on June 19, which is the highest-profile American matchup on the schedule and will likely push transit demand toward the upper end of projections.
The 2026 tournament is not the first World Cup to land on urban transit systems. Three prior editions offer direct benchmarks: Qatar 2022 for free-transit policy, Russia 2018 for high-frequency urban metro performance, and Brazil 2014 for what happens when infrastructure falls short. Each of those tournaments now sits in the NTD and public-data record, making comparison possible.
Doha's 2022 World Cup introduced a model worth studying closely. All transit was free for FIFA ticket holders, integrated into the QR code on the match ticket. The Doha Metro, built for the tournament, recorded over 500,000 trips on Final weekend alone. Mode share to Al Bayt Stadium, Doha's most remote venue, reportedly exceeded 70%, a figure that would have been unimaginable in a car-dependent Gulf city a decade earlier.
Three effects are relevant for 2026. First, free transit dramatically reduced fare-payment friction at turnstiles, cutting post-match egress times. Second, high mode share meant roads near stadiums were manageable despite the absence of large parking facilities. Third, the policy was explicitly designed for international visitors, not just locals: a Brazilian fan landing in Doha did not need to buy a transit card or understand the fare structure. They scanned their ticket and boarded.
Vancouver adopted all three of those principles for 2026. The expected result is the 80% mode share projection embedded in the strain index. TransLink's $21.6M CAD investment is the price of that policy choice. Whether it is the right choice depends on how you value dispersal speed, equity, and the externalities of stadium-area gridlock against foregone fare revenue.
Russia 2018 attracted approximately 3 million visitors. Moscow hosted the Final at Luzhniki Stadium (81,000 capacity). The Moscow Metro, one of the world's deepest and highest-frequency systems, handled the surge with extended hours and minimal service failures. On group stage match days at Luzhniki, ridership at adjacent stations was roughly 40% above comparable non-match days. On Final day, the uplift was closer to 70%.
Two structural differences explain why Moscow absorbed this better than New York will. First, Luzhniki station sits directly under the stadium, with no shuttle required. Second, the Moscow Metro runs at far higher frequency than NJ Transit, with two-to-three-minute headways during peak service and multiple connecting lines feeding the stadium. MetLife, by contrast, relies on a single-track shuttle from a junction station with no parallel capacity.
The Russia 2018 lesson: when transit serves the stadium directly and has genuine redundancy, World Cup demand is manageable even at very large venues. Mexico City and Vancouver both meet that condition. New York, Atlanta, and Seattle only partially meet it.
Brazil 2014 spread matches across 12 cities, many with limited rail. Sao Paulo, Rio de Janeiro, and Fortaleza relied on shuttle buses and temporary services. Post-match dispersal queues of two or more hours were documented at multiple venues. Fan transport complaints ran second only to accommodation in FIFA's post-tournament reports.
The structural parallel to 2026 is the MetLife/Secaucus corridor. One stadium, one access route, no redundancy, high demand concentration. The key difference is that NJT has months of preparation time and an established commuter rail system. Brazil 2014 venues often had none of that. But the physical constraint of a single-corridor stadium is not solved by better planning alone.
The Transit Strain Index is a pre-tournament planning estimate. It takes projected incremental demand (stadium capacity multiplied by transit mode share and fill rate) and divides it by available corridor capacity (serving lines, vehicles per hour, vehicle capacity, 3-hour peak window). A score of 10 means demand equals exactly twice scheduled capacity. A score above 7 means the corridor is being substantially outpaced and the system copes through crush loading, extended egress times, or passengers giving up and walking or taking rideshare. The index will be revised against actual ridership figures as they arrive.
| City / Match | Stadium | Peak Strain | Primary constraint |
|---|---|---|---|
| New York/NJ — Final (Jul 19) | MetLife | 9.5 | Single NJT corridor, no rail redundancy to MetLife |
| Atlanta — Semi-final (Jul 15) | Mercedes-Benz | 9.0 | 28% base mode share spiking on single-day knockout demand |
| New York/NJ — Round of 16 | MetLife | 9.0 | Same corridor, elevated fan intensity vs. group stage |
| New York/NJ — Group stage (avg) | MetLife | 8.2–8.5 | 82,500 capacity × 45% mode share = ~37k incremental riders/match |
| Vancouver — Round of 16 (Jul 7) | BC Place | 8.6 | Post-match SkyTrain crush even at enhanced frequency |
| Seattle — Group stage | Lumen Field | 7.1–7.3 | Constrained waterfront egress, 72k fans in 30-minute window |
| Mexico City — Group stage | Azteca | 7.0–7.2 | Line 2 capacity manageable; late-night demand spikes after evening kickoffs |
| Atlanta — Group stage | Mercedes-Benz | 6.4–6.8 | Low base mode share limits absolute transit rider volume |
| Toronto — Group stage | BMO Field | 6.0–6.3 | Smallest venue (45k) offsets high mode share; compact streetcar corridor |
Transit fares for stadium trips sit at the intersection of pricing policy, urban geography, and political economy. The six-city comparison here reveals gaps that reflect deliberate choices as much as cost structures. Three comparison frames put the numbers in context: absolute cost in USD, cost relative to local workers paid at the minimum wage, and cost relative to the typical international fan traveling to the tournament.
| City | Roundtrip (USD) | WC Special | Min wage/hr | Trip as % wage | % of $175 intl budget | Free? |
|---|---|---|---|---|---|---|
| Mexico City | $0.55 | — | $1.43 | 38% | 0.3% | No |
| Vancouver | $0.00 | — | $12.70 | 0% | 0% | Yes |
| Toronto | $4.84 | — | $12.57 | 38% | 2.8% | No |
| Seattle | $7.00 | — | $16.28 | 43% | 4.0% | No |
| Atlanta | $5.00 | — | $7.25 | 69% | 2.9% | No |
| NYC (standard) | $25.00 | $150 (Secaucus) | $17.00 | 147% | 14.3% | No |
The % of hourly minimum wage metric normalises fare burden by local purchasing power for residents at the wage floor. It reveals that Atlanta's low absolute fare ($5) still represents a large fraction of one hour's minimum wage because Georgia's wage floor is so low. NYC's standard NJT fare exceeds one hour's minimum wage entirely.
The international fan lens changes the picture significantly. Using $175/day as a median estimate for a traveling World Cup visitor (based on FIFA and national tourism authority data from Qatar 2022 and Brazil 2014), virtually all the fares except the NJT special are negligible fractions of a day's budget. The NJT VIP package at $150 is 86% of one day's median budget. For a fan who has already spent thousands on flights and tickets, this is still a painful add-on rather than a prohibitive barrier.
For comparison: Qatar 2022 transit was free for all ticket holders. The Moscow Metro in 2018 cost roughly $0.60 USD for a roundtrip during the tournament. Brazil 2014 fares ranged from $0.80 to $2.50 per trip depending on city. In that context, North America 2026 spans a wider range than any prior World Cup, from free (Vancouver) to $150 (NYC VIP), within the same tournament.
The "Secaucus VIP package" has become the focal point for fare-equity debate in the US press. At approximately $150 versus a standard $25 roundtrip, it bundles guaranteed boarding on reserved trains, dedicated platform access, and shuttle service. Critics argue this creates a two-tier transit experience priced to extract surplus from a captive audience. Defenders note that the standard service runs at standard fares; the premium is optional.
The underlying economics matter here. MetLife Stadium is in the Meadowlands, accessible only by NJT rail or private vehicles. No subway serves it. NJT cannot add unlimited trains. Given fixed supply, allocation via price is one mechanism. The policy question is whether a public authority should use that mechanism, and whether the disclosures are clear enough that fans understand what they are buying and what alternatives exist.
Verify current NJT pricing at njtransit.com. The $150 figure comes from pre-tournament reporting by Front Office Sports and NBC Boston; it is subject to revision.
NJT's premium product captures incremental revenue from World Cup demand. The trade-off is harder to quantify. Academic literature on transit pricing (Glaeser, Kahn, and others) consistently finds that pricing events shape long-term mode choice. An agency that prices fans off trains at a World Cup may be reinforcing the choice to drive to the next stadium event. Whether the revenue justifies that risk depends on how the agency values future ridership against current income.
LA hosts the 2028 Summer Olympics. SoFi Stadium is LA Metro-served. The Airport Metro Connector and Crenshaw/LAX extensions are currently under construction. The MetLife results from this tournament provide advance evidence on three things LA planners care about: how demand distributes when a city lacks direct rail to its main stadium; what service uplift is achievable on four to six weeks' notice; and what happens to roads when transit hits capacity. Atlanta's data on a single-corridor system handling a semi-final offers a more directly comparable case study than any prior event in LA's own history.
The Penn Station to MetLife corridor carries an estimated 35,000 transit riders per match day. Over nine matches (five group stage through the Final), that is roughly 315,000 person-trips through a defined, captive advertising corridor. NJT currently earns effectively zero per-trip advertising revenue on that route. The Stadium-Chinatown SkyTrain segment in Vancouver, scaled to BC Place's seven matches at roughly 41,000 transit riders per match, represents around 285,000 corridor impressions. Both agencies could monetise station branding agreements with tournament sponsors. Neither appears to have done so at scale.
This analysis makes predictions. Tracking whether those predictions hold is both scientifically important and useful for understanding how well the model captures the actual dynamics of transit demand during a major tournament. This section updates as data arrives.
| Prediction | City | Predicted value | Actual value | Error | Data source / status |
|---|---|---|---|---|---|
| MTA match-day ridership uplift (full system) | NYC | +14% on match days | Pending | — | MTA Hourly 5wq4-mkjj, ~4–6 wk lag |
| CDMX Line 2 uplift on match days | CDMX | +12% on match days | Pending | — | datos.cdmx.gob.mx, ~2 day lag |
| NJT Meadowlands boardings per group stage match | NYC | ~35,000 | Pending | — | NJT post-event release |
| Vancouver SkyTrain uplift on BC Place match days | VAN | +16% | Pending | — | TransLink post-event report |
| MetLife peak strain index (group stage) | NYC | 8.2–8.5 | Pending | — | Model estimate; calibrate against NJT boardings |
| Atlanta MARTA boardings per match day | ATL | ~19,000 | Pending | — | NTD monthly release, August 2026 |
| Vancouver mode share on match days | VAN | ~80% | Pending | — | TransLink post-event survey |
Predictions are based on: (a) pre-tournament baseline ridership from 2025 NTD and agency data; (b) uplift factors estimated from comparable mega-events (prior World Cups, major NFL playoffs, Pan Am Games); and (c) corridor capacity parameters documented in the Methods Appendix. An error column populates as actuals arrive. If predictions systematically over- or underestimate, the methodology section will document what drove the deviation.
| Source | Granularity | ID / URL | Lag |
|---|---|---|---|
| FTA NTD Monthly | Agency × month, 2002–present | data.transportation.gov/8bui-9xvu | ~3 months |
| MTA Subway Hourly | Station × hour, 2025– | data.ny.gov/5wq4-mkjj | 4–6 weeks |
| CDMX Metro Afluencia | Daily, by line/station | datos.cdmx.gob.mx | ~2 days |
| GTFS Feeds | Static schedules | Transitland archive | Realtime |
| Fares / Surge Plans | Manual, June 2026 | data/manual/fares.csv | Manual update |
| Match schedule | Match-level | data/manual/matches.csv + fifa.com | Manual update |
The years 2020 and 2021 are excluded from the synthetic control pre-period fitting window by default. They are included with a dummy variable in a robustness specification. The baseline panel uses 2015–2019 as the primary pre-treatment period, with 2022–May 2026 as additional post-recovery data to establish the current ridership trajectory. Any analysis including 2020–21 includes explicit documentation of the choice.