The Price Of A Clear Road

AEvery driver caught in a queue is at once a victim of congestion and a cause of it. The insight is not new — the Norwegian economist Ingeborg Sandvik set it out formally in 1963 — but its policy consequence took forty years to reach a working street. A motorist entering a crowded city pays for fuel, for the vehicle and for a parking space, yet pays nothing for the seconds of delay imposed on everyone queuing behind. Because that cost is real but unbilled, traffic settles at the level where the road is used until it is barely worth using. A charge for entry, Sandvik argued, would not so much punish drivers as return to them information the market had withheld. Engineers of the period preferred to build their way out of the problem, and for three decades they did. The idea was retrieved from the journals only when new urban road capacity became politically as well as financially unaffordable.

BThe first scheme to survive a full political cycle was the cordon introduced in Kestermon in 2003. Vehicles crossing an inner boundary between 07.00 and 18.30 on weekdays paid a flat charge, five units of the local currency at the opening and eight by 2016. The immediate figures were striking: crossings fell by 21 per cent in the first year and average speeds inside the zone rose by about 14 per cent. The figures a decade later were less quotable. By 2014 speeds within the zone had drifted back to roughly their pre-charge level, which opponents read as proof that the charge had failed. The city's transport authority offered a different account. The road space released by the charge had been handed deliberately to bus lanes, cycle tracks and widened pavements, so by 2014 the zone was moving at pre-charge speeds while carrying substantially fewer cars than it had before 2003. Whether that counts as failure depends on what the charge was for.

CVesna Rasmussen, a transport economist at the Brentmoor School of Urban Studies, has compared fourteen schemes operating between 2004 and 2019, and argues that the public debate has fastened on the wrong number. Average speed is what journalists ask for; what hauliers and bus operators respond to is the difference between an ordinary Tuesday and a bad one. Across her sample the mean journey time fell by 9 per cent, while the variability of that journey time fell by 27 per cent. A delivery firm, she points out, does not schedule for the average day but for the worst plausible one, and every minute of that margin is paid for in vehicles and wages. Rasmussen also notes that in eleven of the fourteen cities the money raised was committed by statute to public transport, a design she regards as doing more for acceptance than for efficiency.

DKwame Ofori, an urban geographer at Ardleigh University, is the most cited critic of such schemes, though not for the reason usually attributed to him. A flat charge is regressive in the abstract, since an identical sum is a larger share of a small income; but Ofori's survey of 4,100 Kestermon households found that the poorest fifth made only 4 per cent of charged crossings, most of those households already travelling by bus before the scheme began. His objection concerns geography rather than income. The traffic did not simply evaporate: on the ring of residential streets immediately outside the boundary, vehicle counts rose by 6 per cent, and the people living there receive none of the spending the charge funds. He is unpersuaded, too, that the burden falls evenly among those who do pay. An office worker can leave earlier; a plumber carrying tools to a job booked for ten o'clock cannot, and it is the journeys that cannot be moved which pay the charge most often.

EPublic opinion has followed a strikingly consistent path. When Vallmar ran a seven-month trial in 2007 and then put the scheme to a referendum, approval rose from 34 per cent before the trial to 53 per cent at the vote, a sequence of resistance before and acceptance after that has since been recorded in several other cities. Rasmussen draws a practical lesson from it: a scheme should be tried before it is voted on, because its benefits are experienced while its costs are merely imagined. Ofori accepts the figures and questions what they measure. The Vallmar ballot was open only to residents of the municipality inside the boundary; the commuters from the eleven surrounding municipalities, who paid the charge without receiving the transport spending it financed, were never asked.

FThe cordon may in any case prove a transitional device. Several administrations are preparing distance-based charging, under which a vehicle is billed by the kilometre, by the hour of the day and by the emissions it produces, an arrangement that requires its movements to be recorded. Rasmussen treats this as the logical destination of Sandvik's argument, since a cordon prices a line on a map rather than the delay a driver actually causes. Ofori's caution is not about surveillance alone: a charge per kilometre moves the burden outwards, towards households living furthest from the centre because they could not afford to live in it. Both accept that the fiscal question is closing in, as electric vehicles pay almost no fuel duty and road budgets must be found elsewhere. What separates them is purpose. For Rasmussen the charge should be set at whatever level clears the road; for Ofori a price that is efficient across a city may still be unjust street by street, and a scheme unable to say which streets it is prepared to burden is not yet finished.