The Ground That Pays For Itself

AFew kinds of public spending are argued over as persistently, or settled as unconvincingly, as the money cities put into sports stadiums. The modern pattern dates from the early 1990s, when ageing venues across North America and Europe were replaced by buildings designed as much for retail and hospitality as for the game. The bills are not small. A survey of 118 major stadiums opened between 1992 and 2020 puts the median construction cost at 480 million dollars, of which taxpayers met a median 54 per cent, usually through bonds repaid over thirty years — a term that in several cases outlasts the structure it paid for. The case put to councils is always a version of the same argument: a stadium draws visitors, visitors spend, spending creates jobs, and the jobs return in taxes what the bonds took out. It is intuitive, it comes supported by reports the clubs themselves commission, and for three decades the academic literature has been almost uniformly against it.

BRania Duplessis, a sports economist at the Lindqvist Institute for Urban Policy, has assembled the largest audit of that literature. Her 2021 review pooled 137 studies published between 1979 and 2019, each measuring what happened to employment, wages or personal income across a metropolitan area after a stadium opened. The pooled estimate was indistinguishable from zero, and the handful of clearly positive findings clustered, she notes, among studies paid for by clubs or leagues. Her explanation is not that stadiums destroy value but that they move it. Households keep a rough budget for going out; money spent on a ticket on Saturday is money not spent at a cinema, a restaurant or a bowling alley in the same city on some other evening. Economists call this substitution, and it never appears in the figures a club presents, because those figures count the crowd at the turnstile and never the seats left empty elsewhere.

CTwo further leaks follow. The first is that a stadium's largest single cost, the players' wages, is paid to people who often live and invest outside the city, so the multiplier consultants apply to match-day spending overstates what remains in the region; Duplessis's estimate for a mid-sized city is that about 30 cents of every dollar taken at a fixture has left it within a year. The second leak is quieter and, she argues, the more important of the pair. Public money committed to a stadium is money not committed to something else, and the alternatives — road maintenance, a tram line, a technical college — tend to yield returns that are unglamorous but real. On this reading the honest question is never whether a stadium delivers any benefit at all, but whether it delivers more than the bus depot that went unbuilt.

DHenrik Voss, who teaches urban design at the Marchmont School of Architecture, disputes none of Duplessis's arithmetic and calls her review the best available. His objection is to the unit of measurement. A region of two million people offers, he argues, far too blunt a frame to register what a single building does; asking whether a stadium lifted regional income is like asking whether one new library lifted national literacy. His group tracked the streets instead. Across 24 stadiums opened in mid-sized European cities between 2004 and 2018, ground-floor vacancy within four hundred metres of the site fell from 19 per cent to 8 per cent over the following decade. The improvement, however, was confined to the eleven venues stitched into an existing street grid and kept in use on days without fixtures, for markets, offices or clinics. Where the ground sat inside a ring of surface parking, vacancy did not move.

EThe dispute reaches the ballot box. Between 1995 and 2022, voters in 43 North American cities were asked to approve stadium subsidies directly; 24 of those measures passed, most of them by thin margins, and the public share of construction costs has drifted downwards since the mid-2000s as councils have learned to bargain. Clubs have adjusted their pitch accordingly, leaning less on jobs and more on what economists awkwardly call psychic income — the pleasure a population takes in having a team, whether or not anybody attends. Surveys asking residents what they would pay each year to keep a club in town routinely return figures which, multiplied across a city, would justify a stadium several times over. Duplessis treats those surveys as the weakest evidence in the field, since a respondent parts with nothing when answering. Voss will not go that far, though he accepts that such figures should never be totalled and presented as a sum the city has somehow earned.

FWhat separates the two is narrower than the public argument suggests. Both hold that the job-creation figures in consultants' reports are unusable, and both accept that the money has to come from somewhere. Duplessis's position is that the burden of proof sits with the club, and that three decades of null results should govern how councils receive any promise of growth. Voss's is that those null results describe the average stadium, most of which were built as isolated boxes beside motorways, and that an average is a poor guide to a decision about one site on one street. The matter left open, and it is an empirical one, is whether the eleven promising cases in Voss's sample owed their outcome to the building itself or to the sort of council that would put a stadium in the middle of a working neighbourhood in the first place.