At twenty past eight on a Tuesday morning, the cycle path on the Eerste Constantijn Huygensstraat in Amsterdam West is doing something that the people who built it never planned for: it is queuing. Not gridlocked, exactly, but pulsing — a dense, two-metre-wide column of commuters, a cargo bike with two children strapped in front, a fatbike rider hunched and impatient, a delivery rider on a speed pedelec who would very much like to be somewhere else. Where the path narrows at the junction, the column compresses, then releases. The Dutch have a word for what happens when it does not release. They call it the fietsfile — the bicycle traffic jam.
This is the problem that thirty years of success built. Amsterdam spent a generation winning the argument that a city should be designed around the bicycle, and it won so completely that the bicycle has now run out of room. The municipality’s response, on the evidence of the past three years, is unfashionable. Rather than chase ever-higher throughput, the city has begun, quietly and contentiously, to redesign its busiest cycling space for less speed, more separation, and a slower kind of flow.
The success that became the problem
The numbers vary by who is counting and what they count — trips, movements, commutes — but every credible figure puts the bicycle’s share of Amsterdam journeys somewhere around a third to a half. By the city’s own commuting data, well over half of working residents cycled to work in recent surveys. There are, on a normal day, several hundred thousand bicycle trips across the city. For most of the 2000s and 2010s this was the headline a planner dreams of. The trouble is that a cycling network sized for the Amsterdam of 2005 is now being asked to absorb the Amsterdam of today, and the bicycles themselves have changed.
The decisive shift is the electric motor. A conventional commuter pedals at perhaps fifteen kilometres an hour; an e-bike rider does twenty-five; a derestricted fatbike or a speed pedelec does considerably more. Put all of them on the same strip of red asphalt as a child, a tourist and a grandmother on an omafiets, and the issue is no longer congestion alone but the spread of speeds and masses sharing one lane. The municipality has been blunt about the consequence: on the narrowest and busiest paths, the differences in speed have outgrown the infrastructure, and collisions on some routes have risen.
Building downwards, and under the water
The most visible crisis was never on the road but at the kerb: where to leave the thing once you arrive. Amsterdam Centraal had for two decades been served by the Fietsflat, a three-storey open-air rack on Stationseiland that held around 2,500 bicycles and overflowed perpetually onto every railing within walking distance. It was an emblem of the problem dressed as a solution.
The replacement is genuinely strange and genuinely large. In late January 2023 the city opened the Stationsplein garage — built not underground but underwater, excavated beneath the Open Havenfront, the stretch of water in front of the station — with space for close to 7,000 bicycles. Weeks later the IJboulevard garage opened on the far side of the station, beneath the IJ, adding roughly 4,000 more. Together the two underwater halls take in around 11,000 bicycles, and the first twenty-four hours of parking are free, a deliberate nudge to keep the racks turning over rather than silting up with abandoned bikes. With the new capacity in place, the city resolved to dismantle the old Fietsflat entirely; demolition began in late 2025.
Taking the space from somewhere
None of the surface fixes — wider paths, separate turning lanes, more room to queue — are possible without taking space from something else, and in Amsterdam that something else is almost always the parked car. Since 2019 the municipality has run a deliberate policy of attrition on on-street parking, aiming to remove on the order of 1,500 spaces a year, with a stated target in the range of 10,000 to 11,000 removed by the middle of the decade. The mechanism is less confrontational than the headline: rather than tow cars, the city largely declines to reissue residents’ parking permits as they are surrendered, and reclaims the freed asphalt for trees, footways and bicycle parking.
Divide the public space by speed, not by vehicle type: thirty on the roadway, twenty on the cycle path.
Amsterdam Bike City Innovation Lab, trial design 2024
The slow turn: designing for pause, not throughput
The genuinely new idea is harder to photograph than an underwater garage. For years the orthodoxy — codified nationally in the CROW design manual for bicycle traffic and its principles of cohesion, directness, safety, comfort and attractiveness — treated a cycle path as a flow problem: move the most people with the least friction. Amsterdam’s recent experiments invert that logic on its busiest streets. They accept friction as the price of safety and reach instead for separation.
The first move was old: in 2019 the city pushed mopeds and scooters off some four-fifths of the cycle paths inside the A10 ring and onto the carriageway, with a helmet requirement attached. By the municipality’s own evaluations the paths grew calmer and crashes fell, though compliance proved imperfect and had to be policed. The second move is newer and stranger. Between April and July 2024 the city ran a pilot on the Eerste Constantijn Huygensstraat and the Bilderdijkstraat that did not sort traffic by vehicle but by velocity: anyone travelling above twenty kilometres an hour was invited onto the adjacent thirty-kilometre roadway, leaving the cycle path to the slow. In practice only a minority of fast riders took up the offer — but the cyclists who remained reported, by the city’s account, more calm and more comfort on the path. That is the inversion in one sentence: the metric of success was not how many got through, but how it felt to be there.
The logic is now spreading. A later round of trials in the Vondelpark, the Westerpark and the Rembrandtpark set out simply to coax fast cyclists into riding slower, and the city has signalled a willingness to test a twenty-kilometre limit on the cycle path itself — with, for now, no fines attached. The candidness about the gaps is part of the method. A speed limit nobody enforces is, as critics have said plainly, close to symbolic.
The negotiating table
What makes this difficult is that the conflict is no longer cyclist versus driver, the comfortable Amsterdam fight of the past forty years. It is cyclist versus cyclist, and cyclist versus pedestrian. The commuter on the e-bike, the parent on the cargo bike, the schoolchild on the fatbike and the tourist wheeling a rented city bike all have a legitimate claim on the same red strip, and they disagree about how fast it should move. Every widening takes parking; every speed rule takes liberty from somebody who was, until last year, simply cycling. There is no design that satisfies all of them, which is why the city keeps returning, pilot by pilot, to the negotiating table rather than the drawing board.
It is worth being honest about how provisional all of this is. The underwater garages are built and full; that part is settled infrastructure. The speed experiments are exactly that — experiments, several still running, none yet hardened into the kind of binding standard that turns a trial into a network rule. Whether Amsterdam can codify slowness the way it once codified the cycle path is the open question of the decade.
What the queue is really for
Back on the Huygensstraat the morning column thins by nine. A man stops mid-path, one foot down, to let a child wobble across to school; behind him three riders brake without complaint, because braking is now what the path is for. This is the unglamorous shape of a city that succeeded too well: not a triumphant grid moving ever more bicycles ever faster, but a crowded public room being slowly, awkwardly renegotiated so that everyone in it can still get home.
The rest of cycling Europe — the Copenhagens and Parises and Utrechts now building the infrastructure Amsterdam built a generation ago — will eventually arrive at the same wall. Amsterdam’s contribution to the European urban decade may turn out to be less the bike lane itself than the harder, later lesson: that the reward for filling a network is the obligation to redesign it for pause. The city is, by its own admission, still working out how.
Sources: Gemeente Amsterdam (Meerjarenplan Fiets; Amsterdam Bike City / Innovatielab; municipal evaluations of the snorfiets-naar-de-rijbaan and snelle-fietsers-naar-de-rijbaan trials, 2019–2024); CROW Ontwerpwijzer Fietsverkeer (Design Manual for Bicycle Traffic); reporting by NL Times, DutchNews.nl and Bloomberg on Amsterdam’s parking-reduction policy and the Stationsplein and IJboulevard underwater bicycle garages; Wikipedia “Fietsflat” for the former Stationseiland facility and its demolition.
