The Bike Lane Nobody Can Use

Last week I was at Lake Garda in Italy — one of the most beautiful places you can point a bicycle at anywhere in Europe.

And it was there, on a quiet road above the lake, that I found the perfect bike lane. Perfect, that is, if your goal was to stop anyone from actually cycling on it.

Look at it. A smooth, freshly painted, properly separated path, safely tucked behind a guardrail away from the traffic. Someone spent real money on this.

Then, every forty metres or so, a pair of steel barriers — offset, red-and-white striped, spaced so tightly that I had to slow to walking pace, swing wide, and thread my handlebars through at an angle. Every. Forty. Metres.

Riding it isn't cycling. It's dressage.

Anyone on a cargo bike, or towing a child trailer, or riding a handcycle or an adapted trike, simply can't use it at all. Their bike lane is the road next to the bike lane.

Why the barriers are there

These things have an official name: access controls. Councils — Italian, British, everyone's — install them for one stated reason: to keep motorbikes and mopeds off the path. This one is a shared pedestrian and cycle path, so there's a second, unspoken reason too: to slow cyclists down so they don't frighten people walking.

It sounds sensible in a planning meeting. The evidence says it's nonsense.

In 2016, the city of York removed thirty of these barriers from a single path — the Foss Islands route. Then somebody actually counted what happened next. Over a two-month monitoring period, the path recorded roughly 22,000 legitimate trips for every one unauthorised motorbike or quad trip. Meanwhile, an average of 45 trips a day were made by people using pushchairs, wheelchairs and cargo bikes — the exact people the barriers used to shut out.

When Sustrans asked West Midlands Police whether motorcycle barriers actually reduce antisocial riding, the answer was that there is no evidence that they do. One council in Nuneaton, challenged over the K-frame barriers on its own cycle route, admitted it couldn't point to any misuse on the route at all — and kept the barriers anyway. Which makes intuitive sense, if you've ever watched a determined teenager on a moped: the barrier stops a wheelchair dead and delays the moped by about four seconds.

The UK's own national design guidance, LTN 1/20, has said since 2020 that access controls like these "should not be used" — they exclude disabled riders, cargo bikes and anyone on a non-standard cycle. Sustrans has identified around 16,000 barriers across the National Cycle Network that need removing or redesigning, and is now working through them — at the current pace, the job runs to 2040.

Read that again: Britain built so many obstacles on its own cycle network that removing them is a twenty-year national infrastructure project.

And it matters most for the people with the least choice. In Wheels for Wellbeing's survey of disabled cyclists, over 75% reported difficulty accessing cycling, with inaccessible infrastructure the top barrier — ahead of even the cost of adaptive bikes. A third had been asked to dismount while using their cycle as a mobility aid. This July, just up the road from where I was riding, an Italian woman on a mobility scooter was forced off a cycle path in Carpi and into live traffic by exactly the same style of barrier.

A very British disease too

I don't want you thinking this is an Italian problem. Britain practically invented the unusable bike lane.

The Warrington Cycle Campaign has run a "Facility of the Month" feature since 2001 — a quarter of a century of lanes that end in lamp posts, lanes painted through bus shelters, and cycle routes with "cyclists dismount" as a design philosophy. It became a book, honestly titled Crap Cycle Lanes. There was enough material for a sequel.

My favourite entry in the genre: Britain's shortest cycle lane, in Stroud, Gloucestershire — around 2.4 metres long. You could overshoot it by sneezing.

Funny, until you see the invoices.

In 2020, London's Kensington High Street got a protected cycle lane costing around £700,000. The council removed it seven weeks later, cutting short its own 18-month trial before the data could embarrass anyone. Then, in 2023, after a consultation found 66% support for cycle lanes, the same council put some back — as painted dashed lines. Protected lane, no lane, paint. Three positions in three years, on one street.

Middlesbrough went one better. The Linthorpe Road cycle lane cost £1.7 million to build in 2022. In 2025 the council started ripping it out, and the removal cost £2.7 million — a million pounds more than the lane cost in the first place. Here's the detail that should sting: the council's own figures, released under Freedom of Information, showed that after the lane went in, cycling on the road rose 70% and cyclists' share of injuries halved, from 23% to 12%. It was working. They removed it anyway.

That's £4.4 million of public money to end up exactly where they started, minus the safety improvement.

The strange tale of the €1.2 billion bike path

Back to Lake Garda, because the lake with the barrier-slalom path is also home to the most expensive bicycle project in the world — and the contrast tells you everything.

The Ciclovia del Garda is a plan to build a cycle route around the entire lake: 165 kilometres, across three regions. When it was approved in 2019, the budget was €344 million — already vast. By this spring, local authorities were warning the true cost could reach €1.2 billion. One 5.5-kilometre stretch between Limone and Riva del Garda is now priced at €95 million. That is roughly €17,000 per metre of bike path. Italy's Court of Auditors looked at the costs and used the phrase "grossly disproportionate" — the national average is nearer €320 per metre.

In December, Lombardy quietly gave up on the final 27 kilometres of its shore altogether. The replacement for that section of the world's most expensive bike path? A ferry.

And yet — the first piece of the Ciclovia, the cantilevered path bolted to the cliff face at Limone sul Garda, is genuinely magnificent. About two kilometres, roughly €4.7 million, suspended over the water, and so beautiful that it's clocked over two million visitors and a reputation as the most beautiful bike path on Earth. I recommend it without reservation.

But notice what got built and what didn't. The spectacular two kilometres for tourists: delivered, celebrated, photographed daily. The ordinary, useful network for people who just need to get to work and school: over budget, behind schedule, partly cancelled — and where it does exist, sliced up every forty metres by steel barriers.

Spot the family with the child trailer. They're on the road — next to a lakeside path they can't use.

What the story actually is

It would be easy to file all this under incompetence. I don't think that's quite right.

Every example above is the fossil of a political compromise. One administration wants a bike lane; the opposition wants it gone; local businesses want parking; somebody's worried about mopeds; a pedestrian group is worried about speeding cyclists. Nobody wins the argument, so the infrastructure itself becomes the argument — a protected lane one year, tarmac the next, paint the year after. A path is built, then barriered into uselessness as a peace offering to its opponents.

A bike lane, though, is a long-term bet. It changes behaviour over years — the Middlesbrough numbers prove that — while elections happen every four or five. Build-remove-rebuild isn't a transport strategy; it's the election cycle drawn in thermoplastic paint.

And the maths of compromise is brutal, because a bike lane is a network good. A lane you can ride 90% of — the other 10% blocked by barriers, or missing, or ending in a lamp post — isn't 90% as useful. For the person on the cargo bike, the parent with the trailer, the rider on the handcycle, it's 0% as useful. Half a bike lane isn't half as good. It's a press release with a maintenance bill.

The places that got this right — Paris most famously, where bikes now outnumber cars on many central streets at rush hour — didn't build anything spectacular. They built boring, continuous, direct routes and then refused to rip them out when the grumbling started. That's the entire trick. Not cantilevers, not LiDAR, not €17,000 a metre. Continuity, and nerve.

No barriers, no cantilevers — just somewhere pleasant to ride. It isn't complicated.

So here's my modest proposal, offered free of charge to any transport minister, assessore or borough councillor reading: before you announce the next bike lane, ride the last one you built. All of it. On a cargo bike.

If you can get through without dismounting, congratulations — you may build another.

Viva la Velorution,
Pete