In July 2026, the largest FIFA World Cup ever staged came to a close across the United States, Canada and Mexico. Forty-eight teams, 104 matches, 16 host cities — as the tournament grew, so did the numbers attached to it [source: FIFA, 2026]. Ahead of kickoff, an analysis commissioned by FIFA and the World Trade Organization (WTO) projected that the event would add $40.9 billion to global GDP and create about 824,000 jobs [source: FIFA, 2025]. Dazzling figures.
Yet economists have been asking the same question for decades. Do mega-events — the giant international tournaments like the World Cup and the Olympics — actually pay off for the countries that host them? What is striking is how one-sided the academic answer has become. This piece is not about who won a given match; it is about the general economics of mega-events. The key is to keep two distinctions straight: who produced a number (organizer and consulting estimates versus independent research), and when it was produced (pre-event forecasts versus post-event measurements).
Three kinds of number appear below, and they are worth naming before they start arriving. The first is peer-reviewed research — economists measuring what happened after an event. The second is organizer and consulting forecasts, produced beforehand by parties with a stake in the answer. The third is reported realized cost: the running totals news organizations and national auditors assemble once the bills come in, which often disagree with one another because they draw the boundary around "the event" in different places. Nothing that follows upgrades a forecast into a measurement, or a press estimate into an audit.
The 1994 USA World Cup was sold on a promised $4 billion gain. After it ended, academic analysis calculated something else entirely: a $5.5–9.3 billion loss across the nine host cities.
Table of contents
- Why now: the biggest tournament ever, and an old question
- The tournament that changed shape
- Two kinds of number on the same scale
- How the rosy numbers get built
- The official 2026 forecast
- Four reasons the numbers grow
- The winner's curse and what it costs
- The gap between forecast and reality
- The lesson called 1994
- The Olympics that never came in on budget
- What 130 studies say together
- The bill and the "white elephants"
- Brazil 2014 and the arena in the rainforest
- Athens 2004 and the debt that followed
- Sochi and Tokyo: the records for scale
- Qatar 2022: the number everyone misreads
- Will 2026 be different?
- The one structural advantage
- Why "existing" does not mean "free"
- Why cities host even when it doesn't pay
- Three kinds of benefit, and which two get overstated
- Prestige, soft power, and the honest verdict
- Conclusion: what to watch
- What the evidence says so far
- The three questions to ask of any number
Why now: the biggest tournament ever, and an old question
The tournament that changed shape
The 2026 World Cup was record-breaking in several senses. The field grew from 32 teams to 48, the schedule swelled to 104 matches, and for the first time three countries shared the hosting [source: FIFA, 2026]. As the tournament grew, so did the economic-impact estimates attached to it, and the question "how much did this city make?" repeated in host after host. The biggest tournament ever is, for that reason, a useful lens through which to weigh the economics of mega-events again.
The scale is worth stating precisely, because nearly every impact figure is a function of it. The 104 matches broke down into 72 in the group stage and 24 in the knockout rounds, up from 64 and 32. The 16 host cities were split 11 in the United States, three in Mexico and two in Canada, and the tournament ran from 11 June to 19 July 2026 [source: FIFA, 2026]. More matches, in more cities, over more weeks mechanically enlarges every spending estimate built on top of them.
Two kinds of number on the same scale
The trouble is that the scale has two different kinds of weights on it. On one side sit the "forecasts" that organizing bodies and consulting firms release before the event; on the other, the "measurements" economists take after it. The two often diverge. So this piece tries not to mix impressions with data — because the feeling of "full stadiums and crowded fan festivals" does not, by itself, amount to the measurement of "net growth" [source: NC State University, 2026].
There is a second reason to keep the two apart: they are produced by different people for different purposes. A forecast is commissioned to support a decision usually already made, and published while the event still needs public money and public enthusiasm. A measurement appears years later, when nobody is selling anything. The literature also notes that the balance tilts further depending on who hosts: the cost-benefit picture is substantially worse for developing-country hosts, which is precisely where several of the most expensive recent tournaments have been staged [source: Journal of Economic Perspectives, 2016].
How the rosy numbers get built
A mega-event's economic impact usually arrives long before the event itself, in the form of an impressive press release. Understanding how those numbers are built shows why economists grow cautious in front of them.
The official 2026 forecast
The official forecast for 2026 came from an analysis that FIFA and the WTO commissioned from the consultancy OpenEconomics, released in April 2025. It estimated that the 2026 World Cup would lift global GDP by $40.9 billion, create roughly 824,000 full-time-equivalent (FTE) jobs, and draw 6.5 million spectators across the three host countries. For the United States alone, the figures were a $17.2 billion GDP contribution and about 185,000 jobs [source: FIFA, 2025]. The widely cited "$47 billion" figure, however, calls for care: it is the "US output" of the 2026 World Cup bundled together with the 2025 FIFA Club World Cup, also hosted in the US — not the World Cup on its own [source: FIFA, 2025]. Separately, the consultancy Tourism Economics projected about 1.24 million international visitors to the US for the tournament [source: Tourism Economics, 2025].
Two further figures from the same study explain why the headline is so easy to garble. For the United States, alongside the $17.2 billion GDP contribution, the study projected $30.5 billion in gross output — a broader measure that counts total transactions rather than value added, so the two are not interchangeable. And the Club World Cup folded into the $47 billion headline was itself credited with $21.1 billion in global GDP, $9.6 billion in US GDP and 105,000 US jobs [source: FIFA, 2025]. Every one of these is an ex-ante projection prepared for the organizer, not a measurement of anything that has happened.
Four reasons the numbers grow
Economists tend to see such forecasts less as "wrong" than as "structured to be large," and for good reason. Pre-event impact studies typically total up event-related spending and multiply it by a large multiplier (a coefficient assuming one round of spending ripples into several times its value) — and in that process, four things routinely go missing [source: International Journal of Sport Finance, 2009].
- The substitution effect: much of the money spent on the event is money that would have been spent locally anyway, merely rearranged. If it displaces ordinary tourism, dining and leisure spending, it is not "new" money.
- Crowding out: congestion, price spikes and heightened security deter the regular tourists and business travelers who would otherwise have come. Pre-event studies do a poor job of capturing this displacement.
- Leakage: a large share of the revenue flows out of the host economy. The portion going to FIFA or the IOC, to multinational sponsors, and to non-local contractors and workers does not stay in the local economy.
- Inflated multipliers: promotional studies apply large multipliers, but post-event academic work generally finds them overstated. The misuse of multipliers is singled out as a primary reason impact studies overstate the true gains [source: International Journal of Sport Finance, 2009].
Leakage is the one host cities feel most directly. The independent economist reviewing the 2026 figures puts it bluntly: host cities and other public entities generally receive little or none of the direct game-day revenue, because the governing body captures most of the gains through media rights, sponsorships and ticketing [source: NC State University, 2026]. The same review argues that impact studies are "structured in ways that predictably produce large numbers" and "often fail to fully account for public hosting costs." The problem is less that any single input is wrong than that the accounting is asymmetric — benefits totalled generously, costs recorded thinly.
The winner's curse and what it costs
One more factor sits on top: the bidding contest itself pushes hosts toward the most optimistic promise — the "winner's curse." Winning means promising more, and the public money poured in that way cannot be spent where it might have earned a higher return. That is the opportunity cost.
That opportunity cost is where the broader literature lands hardest. A comprehensive survey of the field concludes that large public subsidies for sports facilities are not justified as worthwhile public investments — a judgment that builds on a foundational 1999 finding of no significant positive effect on per-capita income in cities hosting professional franchises [source: Journal of Economic Surveys, 2023]. Because hosting rights go to whoever promises the most, the process systematically selects the most optimistic bid — and it is that bid, not a median one, that becomes the public commitment.
The gap between forecast and reality
Set forecast and measurement side by side, and the most consistent feature of mega-event economics appears: the gap between the two.
The lesson called 1994
The clearest case is the 1994 USA World Cup. At bid time, boosters promised a gain of roughly $4 billion. After the tournament, peer-reviewed analysis found the opposite. Income growth in the nine host cities came in lower than predicted, showing up as a cumulative loss of $5.5–9.3 billion [source: Regional Studies, 2004]. Because the promise and the measurement carried opposite signs, this case is often cited as emblematic of mega-event economics.
What makes 1994 the cleanest case is not the size of the gap but the quality of the counting. Rather than tallying receipts, the analysis measured income growth in the nine host cities against what had been predicted for them and found it fell short — the comparison a promotional forecast never runs on itself [source: Regional Studies, 2004]. It is also the best-documented instance of predicted tourism gains failing to materialize once displacement is netted out: visitors did arrive, but partly instead of other visitors, and in numbers that did not cover what the host cities had put in.
The Olympics that never came in on budget
Turn to the Olympics and another regularity appears. According to the Oxford Olympics Study, every Games without exception ran over budget, with an average overrun of 156% in real terms (median 90%) — the highest of any megaproject type [source: Oxford Olympics Study, 2016]. The 2024 update likewise called the Olympics "the only project type that never delivered on budget." That said, the 156% figure has drawn methodological challenges, so it is more accurate to treat it as the Oxford study's estimate rather than a settled fact [source: Oxford Olympics Study, 2024].
The 2024 update adds two things worth noting. Paris 2024, presented at the time as a restrained Games, still came in at $8.7 billion in 2022 prices with a 115% real-terms overrun — below the long-run average, but an overrun all the same. And the study reports that Olympic costs are now statistically significantly increasing rather than flattening out [source: Oxford Olympics Study, 2024]. The methodological dispute over the 156% figure is real and should stay in view, but it turns on how large the average overrun is, not on whether overruns are normal. On that narrower question there is no disagreement.
What 130 studies say together
Beyond any single event, a survey synthesizing 130-plus studies over three decades on how professional sports teams and venues affect local economies reaches a similar conclusion. The impact is very limited, and even after adding intangibles such as civic pride, the benefits fall well short of covering public outlays; large public subsidies are hard to justify as worthwhile public investments [source: Journal of Economic Surveys, 2023]. A leading review of the Olympics puts it plainly: in most cases the Games are a money-losing proposition for host cities, yielding positive net benefits only under very specific and unusual circumstances [source: Journal of Economic Perspectives, 2016]. The verdict that hosting the World Cup or Olympics brings "no net economic gains" runs in the same direction [source: Brookings Institution Press, 2015].
Two features of that survey are easy to skip past. It covers more than 130 studies across more than three decades, so it is not a lone result a later paper might overturn; and it reaches its conclusion even after crediting the intangible benefits — civic pride, quality-of-life effects — that boosters usually accuse economists of ignoring [source: Journal of Economic Surveys, 2023]. The distributional point is separate. The finding of no net economic gains comes paired with the observation that the wealthy may profit from hosting while middle- and lower-income residents do not, so an aggregate near zero can still conceal a transfer [source: Brookings Institution Press, 2015].
The bill and the "white elephants"
When forecasts miss, what remains is the bill. And the bill often comes with a "white elephant" — a facility that costs a fortune to maintain and is barely used.
Brazil 2014 and the arena in the rainforest
Brazil's 2014 World Cup is the textbook example. Its total cost is generally put at around $15 billion, with private investment accounting for less than 15% — most of it came from public budgets [source: Time, 2014]. The symbol of it all was the Arena da Amazônia, built in the Amazon rainforest city of Manaus. It cost about $300 million but hosted just four World Cup matches, in a city where even the top local club draws crowds in the low thousands [source: Sports Illustrated, 2016].
The $15 billion is a convenient midpoint rather than a settled total. Depending on what counts as tournament spending, the reported range runs from about $11.6 billion up to $20 billion [source: Time, 2014]. That ambiguity matters: when the boundary around "the event" is loose, the same tournament can be defended as modest or attacked as ruinous without anyone misquoting a figure. The public share is not ambiguous. With private investment under 15%, the overwhelming majority of whatever the final total was came from Brazilian public budgets — which is why the stadium legacy became a political question rather than an accounting one [source: Time, 2014].
Athens 2004 and the debt that followed
The Olympic cases are no different. The 2004 Athens Olympics cost about $11 billion, roughly double the original budget, and many of its venues sat abandoned afterward [source: Time, 2012]. The IOC president at the time said that some 2–3% of Greece's external debt could be attributed to the Games — though it is worth noting this is an attributed estimate, and the causal link to the debt crisis itself is disputed [source: VOA, 2012].
Two details fill in that picture. Security alone cost more than $1.2 billion, a line item that barely registered at earlier Games and has been unavoidable since [source: Time, 2012]. Greece's own prime minister attributed the country's fiscal position to decades of mismanagement rather than to a single tournament [source: VOA, 2012]. The honest reading is that Athens was a large, badly timed expense inside a much larger fiscal problem — and that the 2–3% share is an estimate offered by the IOC's own president, not an audited allocation.
Sochi and Tokyo: the records for scale
The records for sheer scale sit on the same trend line. The 2014 Sochi Winter Olympics, at about $50 billion, remain the most expensive Olympics ever [source: Washington Post, 2014]. The Tokyo 2020 Olympics (held in 2021) cost $13 billion by the organizing committee's official count — about double the $7.3 billion offered at bid time. Auditors have suggested the true total ran to $22–28 billion, again exposing the gap between official figures and measured reality [source: France 24, 2022]. It is worth reading the big Sochi and Tokyo numbers as including broad infrastructure, not just venue costs.
Both figures look different next to their bids. Russia's 2007 bid for Sochi put the cost at $12 billion, so the roughly $50 billion realized figure represents about a fourfold escalation — and it exceeded the far larger Beijing 2008 Summer Games, generally put near $40 billion [source: Washington Post, 2014]. Tokyo's overrun has an unusual component of its own: the one-year postponement alone cost about $2.8 billion, and barring spectators removed roughly $800 million in ticket revenue [source: France 24, 2022]. Neither of those was a construction overrun, which is a reminder that "cost" for a mega-event includes the risk of the world changing between the bid and the opening ceremony.
Qatar 2022: the number everyone misreads
Qatar 2022, dubbed the "$200 billion World Cup," is an especially easy case to misread. Qatar's total spend is often cited as more than $220 billion, but only about $8–10 billion of that went to stadiums and tournament operations. The vast remainder was national infrastructure — a metro, a new airport, roads, hotels — long-term development spending for which Qatar used the Cup as a catalyst [source: Forbes, 2022]. To quote the "$200 billion" figure as if it were the cost of stadiums is to distort the facts.
There is a further wrinkle: no single audited total for Qatar 2022 exists. The commonly cited range of roughly $220–229 billion covers spending from 2010 to 2022 across a national development programme, and it is assembled from outside estimates rather than published as one reconciled account [source: Forbes, 2022]. This is the opposite failure mode from an Olympic cost overrun. There, a specific organizing committee blew past a specific budget; here, "the cost of the World Cup" was never a bounded quantity in the first place — which is exactly what makes the headline number so quotable and so misleading.
Will 2026 be different?
The one structural advantage
So will the 2026 World Cup break from this old pattern? There is one structural advantage. The tournament relies mostly on existing NFL and other large stadiums. That means it largely avoids the biggest and riskiest cost that sank many past World Cups and Olympics: building new venues [source: NC State University, 2026].
It is worth being concrete about what that avoids. The most expensive cases in this article share a single line item: purpose-built venues in places with limited post-event demand for them. Brazil built the Arena da Amazônia in Manaus; Athens built a set of venues that later sat abandoned; Sochi assembled an entire Olympic cluster largely from scratch. The 2026 tournament did not have to run that risk, because the stadiums already existed and would have gone on existing, and being used, whether or not FIFA came [source: NC State University, 2026]. That is a genuine structural difference, and the strongest argument any recent host has had.
Why "existing" does not mean "free"
The same expert, however, immediately adds a caveat: those existing stadiums are not "free." Many were publicly subsidized in the first place, and the US host cities staging the event are estimated to face public costs of $100–200 million or more per city for security, transport and fan events [source: NC State University, 2026]. In cities that already draw crowds, such as New York or Miami, the substitution and crowding-out effects bite especially hard, because hotels that would have filled anyway are simply being counted toward the event. In other words, the risk in 2026 is lower than in the past — but not zero.
Set the two halves of that ledger side by side and the shape of the 2026 risk becomes clear. Host cities carry the security, transport and fan-event costs, while the direct game-day revenue largely flows past them to the governing body and its commercial partners [source: NC State University, 2026]. Whether the tournament paid for any given city therefore depends on visitor spending that is genuinely additional, which is exactly the quantity substitution and crowding out erode — and that quantity is not knowable yet. What exists so far are the forecasts; the independent measurements come later, as they did for 1994.
Why cities host even when it doesn't pay
Three kinds of benefit, and which two get overstated
If the economics is so consistently skeptical, why do countries still line up to host? Much of the answer lies in value that never shows up on a balance sheet. A leading review divides the benefits economists weigh into three kinds: short-run tourist spending during the event, a long-run "legacy," and intangible "feel-good" effects such as pride. The first two are routinely overstated; the third is real [source: Journal of Economic Perspectives, 2016]. But that "feel-good" is small and fleeting. A study of the 2014 World Cup found that football results affect subjective well-being "but only briefly" [source: Frontiers in Psychology, 2015].
The asymmetry is worth spelling out: tourist spending and legacy are the two categories that can be put in a spreadsheet, and the two the literature finds routinely overstated. The feel-good effect is the category economists are most willing to grant as real and least able to price [source: Journal of Economic Perspectives, 2016]. That is an uncomfortable combination: the most defensible benefit is also the one that cannot be set against a construction invoice. And the well-being study found the effect present but brief, which limits how much of a permanent deficit a temporary mood can offset [source: Frontiers in Psychology, 2015].
Prestige, soft power, and the honest verdict
Added to this is the motive of national image and diplomacy — so-called soft power. Such value never shows up in GDP, yet it is real. Sometimes a tournament pulls forward transport or airport upgrades a city genuinely needed. Barcelona 1992, one of the rare cases economists count as a "success," is the classic example — though even there, the consensus is that urban renewal, rather than the sport itself, drove the result. The honest conclusion, in short, is not "always a scam" but this: the "economic" case for mega-events is usually weak, yet the reasons countries host anyway — national prestige, soft power, intangible pride — can be legitimate motives in their own right, so long as they are not dressed up as a money-making investment.
Barcelona is the example everyone reaches for, and it deserves careful handling. The consensus is not that the 1992 Olympics generated a return; it is that the city had a coherent urban-renewal programme and used the Games as a deadline and a funding mechanism for it. The sport was the occasion, not the cause. That distinction turns Barcelona from a template into a condition: a mega-event can accelerate infrastructure a city already needed and had already decided to build, and it can buy a kind of visibility no advertising budget would. Neither of those is an economic return, and the trouble starts when they are presented as one.
Conclusion: what to watch
What the evidence says so far
The real question left by the biggest World Cup ever lies not on the scoreboard but in the ledger. The evidence so far points one way. Pre-event forecasts tend to be built to come out large, and post-event measurements — once substitution, crowding out, leakage and opportunity cost are subtracted — routinely fall short of them. 2026 differs from the past in having lowered its structural risk by reusing existing stadiums, but whether that exception is enough to overturn the rule is not yet known.
None of this collapses into a single verdict, and it should not. The independent research is consistent about direction and candid about its limits: measured effects are small, issued forecasts are large, and the difference is structural rather than accidental. The strongest general statement the literature makes is not that hosting always loses money but that positive net benefits appear only under very specific and unusual circumstances [source: Journal of Economic Perspectives, 2016]. That is an exception clause, not a general case — and it is why the burden of proof belongs with the forecast rather than with the economist asking to see the receipts.
The three questions to ask of any number
What to watch from here is clear: how far independent studies, once the tournament is over, confirm or contradict FIFA's $40.9 billion forecast; what the US host cities' public costs actually settle at; and whether subsequent mega-events, such as the 2028 Los Angeles Olympics, carry on the model of controlling costs by reusing existing facilities. The answer to whether a "bigger tournament" was a "better deal" will come, in the end, not from a dazzling forecast but from the quiet reckoning that follows.
There is also a way to read whatever numbers arrive next, and it takes no economics training. Ask who produced the figure — an organizer, a consultancy working for one, or an independent researcher with nothing riding on the answer. Ask when it was produced — before the event, when it is a projection, or after, when it is a measurement. Ask what is inside it — stadiums and operations only, or a decade of national infrastructure folded in. Those three questions separate almost every misleading mega-event number in this article from the defensible one standing next to it.