Transport for London has a number that should alarm every urban planner: a 10% fall in bus speed costs 6% of ridership. Imperial College London found the same pattern across 15 cities. Buses are slowing because the traffic around them is thickening, and the standard remedy — traffic signal priority — has a hard ceiling nobody wants to discuss. That ceiling is arithmetic. On a route running a bus every two minutes, signals are asked to intervene 30 times per hour in each direction. The junction cannot recover between calls before the next bus arrives. TfL's own language is revealing: signals will give buses priority "where possible." Bristol city council's engineers confirmed the worst case — in the most congested corridors, bus priority delivered zero journey-time saving. The economic damage is larger than a transport story usually gets credit for. What holds a city back is the plain difficulty of traversing it. A firm recruits from a smaller labour shed, a worker reaches fewer jobs, a high street draws from a narrower catchment. Much of British high-street decline is, at root, a decline in how easily people can reach it. This is not an amenity issue. It is a productivity constraint. David Andrews, chair of the Light Rail Transit Association, lays out the structural fix: carry the same passengers in fewer vehicles. One tram replaces three or four buses. Fewer signal interventions per hour means priority actually works, and cross-traffic still gets its turn. About 30 French cities have built trams through their centres since Nantes started in 1985, rebuilding their central streets around them. Sheffield's trams share a shopping street with cars and run the same timetable at peak and midday. A second letter, from V Narayanan in London, introduces a sharper political claim: that cycle lanes have cannibalised bus lanes on narrow streets, and cycle infrastructure sits empty outside two brief peak windows. The argument is that mass transit carrying hundreds per hour lost road space to individual transport carrying dozens. Whether the utilisation data bears this out citywide is debatable, but the underlying resource-allocation question is real: when road width is fixed, every metre given to one mode is taken from another. The deeper structural failure is political. Signal priority is cheap, cycle lanes are visible, trams require decade-long capital commitments that span multiple electoral cycles. Politicians optimise for the intervention that photographs well before the next election, not the infrastructure that transforms a city over 20 years. The French example is instructive precisely because it required sustained political will across administrations. The pattern is familiar across British infrastructure: preference for low-capital, high-visibility interventions over high-capital, high-return ones. Bus priority software instead of dedicated tramways. Cycle lanes painted on existing roads instead of grade-separated transit corridors. Each choice is individually defensible and collectively insufficient. The bus speed data is the receipt.
