SILENCE FILLS the official merchandise shop in Tokyo’s Aqua City mall, just steps from the Olympic village. It had hoped to do brisk business selling everything from T-shirts to traditional daruma dolls to hordes of fans. But the pandemic means that the games, which are due to open on July 23rd, will proceed with little fanfare. Sales are just 10% of what was projected, the shop’s manager laments. The outlook for the wider gains from the Olympics is similarly gloomy—but covid-19 is only partly to blame.

Economists have long argued that, rather than consumption, tourism and prestige, the games leave high debt, wasteful infrastructure and onerous maintenance obligations. In a 2016 paper Victor Matheson and Robert Baade, two American academics, concluded that “in most cases the Olympics are a money-losing proposition for host cities.” The games this year offer plenty more grist for the sceptics’ mill. “It would have been better not to have them,” says Suehiro Toru of Daiwa Securities, an investment bank, expressing a sentiment common in Japan among economists and non-economists alike.

Calculating the economic impact of mega-events like the Olympics can be tricky. Organisers and critics argue over which costs are actually incurred by the games, and which are investments cities would have undertaken anyway. But one near-certainty is that the games blow the budget. Alexander Budzier, Bent Flyvbjerg and Daniel Lunn of Oxford University find that every Olympics since 1960 has overspent, by an average of 172% in real terms. (The International Olympic Committee disputes their findings.) Tokyo fits the pattern. In 2013 the price-tag for the games was $7.5bn. By late 2019 the official budget had risen to $12.6bn, and Japan’s audit board reckoned that the true cost was twice that. Covid-19 countermeasures, including the cost of testing and adapting venues, have since added another $2.8bn.

A Tokyo government study in 2017 projected that the returns would more than make up for the costs. It estimated that the games would generate ¥14trn ($127bn) of additional demand. Part of the boost comes from the construction of new infrastructure, though such projects can often crowd out investments in other, more useful areas. Another benefit usually comes from consumption around the games, on everything from tickets to food and drink. But as the competition will unfold without spectators, Tokyo stands to reap little of that.

Much of the gain is expected to come from the woollier category of “legacy effects”, such as increases in tourism and also the use of transport and other infrastructure after the tournament ends. But those projections are probably overstated to begin with, and the pandemic will erode much of them, argues Miyamoto Katsuhiro of Kansai University. Japan may still benefit from the extra hotels that it built, reckons Kiuchi Takahide of the Nomura Research Institute; the supply of rooms was becoming a constraint on Japan’s growing tourism industry before the pandemic. But others worry that the controversy around holding the games during a pandemic will create a negative legacy effect, denting Japan’s international standing and making travellers less likely to visit. Such concerns might explain why big sponsors such as Toyota say they will shun the opening ceremony.

The number of cities bidding for the games has dwindled in recent years. Tokyo’s experience seems set only to put more of them off, especially rich places with less need for infrastructure binges. The Olympics may become a race no city wants to run.