The City That Does Not Close

AThe phrase 'night-time economy' entered planning documents in the early 1990s, when councils in Britain and the Netherlands began to describe the hours after six in the evening as an under-used asset rather than a policing problem. Since then the term has hardened into a policy category with budgets attached to it. In a 2023 survey of forty European cities, work beginning after six in the evening accounted for just under a fifth of all employment, and the largest single group within it was not bar staff but the people who move freight, clean offices and staff hospital wards. That breadth is regularly lost in public argument, which treats the night-time economy as a synonym for late drinking. Two framings compete for the same evidence. To its advocates, the extended evening is a regeneration tool costing a council little more than a signature on a licence. To its critics, it is an accounting device: the takings are private, while the bill arrives at the ambulance station, the housing block and the street-cleaning depot.

BNasir Marchand, an urban economist at the Calderfield Institute of Urban Policy, has spent several years asking what those takings actually consist of. Between 2015 and 2022 her team followed eighteen districts across eleven cities that had extended their licensing hours, matching each against a district in the same city that had not. Employment between six in the evening and six in the morning rose by 14 per cent in the extended districts. Marchand is careful about her own figure: it counts posts rather than hours, and a night post is more often part-time than a day one. Four in five of the additional posts were created in premises licensed for more than three hundred people. Her central finding is less comfortable still. Of the additional late-evening turnover, she attributes 62 per cent to displacement, meaning trade that had merely moved from earlier in the same evening, which leaves a real gain far smaller than the one councils quote. The night, on this account, is less an engine than a timetable.

CThat estimate yields a prediction, and Marchand has tested it. If much of the late trade is trade shifted later, it should appear only where people can still get home once it ends. In 2022 her group compared districts served by a night rail or metro service with matched districts that were not: late turnover rose by 19 per cent in the first group, against 4 per cent in the second. The advantage was conditional in two ways she stresses. It appeared only where the night service ran on four nights a week or more, since a single Saturday-night service made no measurable difference. And it disappeared beyond four hundred metres of a station, whereas the noise complaints generated by the same premises were logged from addresses up to nine hundred metres away. The benefit, in short, is collected within a few minutes' walk of the platform, while the nuisance travels considerably further into the surrounding streets.

DJonas Wieland, a public-health researcher at the Nordhavn School of Population Health, works on the other side of that ledger. Examining the same eighteen districts between 2016 and 2022, he found that ambulance callouts between midnight and four in the morning rose by 27 per cent where hours had been extended, against 6 per cent elsewhere. His argument is not that the trade should be stopped but that it is mispriced: a licence permits a firm to generate a cost which is then met by services the firm does not pay for. Wieland volunteers two limitations. He counted incidents recorded, not incidents that occurred, and the councils that extended hours were in those same years putting more officers on the street, so part of the rise may be closer observation rather than more harm. He also counted callouts without weighting them for severity, so a night of minor injuries can outweigh a quiet week containing one grave one.

EThe dispute now has a legal edge. A licensing code adopted by the city of Dunmore in 2021 requires premises trading after one in the morning to contribute to street cleaning and late transport, and to publish the sums involved. Wieland regards the levy as the most valuable feature of such codes, because it places at least part of the cost on the firms that generate it. Marchand is here the more sceptical of the two. A charge set per premises rather than per customer, she notes, asks least of rooms licensed for several hundred drinkers and most of the small bars, so it presses hardest on the venues least able to pay. Both object to a habit common among councils: quoting headline turnover for a district that has lengthened its hours without first deducting the trade which has merely arrived from an earlier hour of the same night.

FWhat is emerging is a division of the question rather than a verdict on it. Where night activity is anchored to something that cannot be rescheduled, such as a concert hall with a fixed curtain, a container port, a hospital's night shift or a university library, the hours are genuinely added and the spending around them is money that would not otherwise have been spent. Where the night consists of the same drinking arranged later, the ledger largely rearranges itself: the takings move down the clock and the costs move outdoors. Whether cities are drifting towards the anchored kind of night work or away from it is disputed, and neither researcher claims to know. What both reject is the framing that has governed public argument, a lively city set against a safe one, since the streets that perform best on Marchand's measure and worst on Wieland's are frequently the same few.