Good day my good friend.
As I mentioned in last week’s newsletter (this week’s as I write it, which is weird), this newsletter has been scheduled in advance as I have a lot on, including researching something that I hope to publish next week. Hopefully the world hasn’t fallen apart in the meantime.
One thing I did want to flag. As part of the Transport Box project that I am helping to run with the excellent people at UK Unplugged, we are running a transport-related activity session for children with Special Educational Needs or are on the CAMHS waiting list in Stockport this Wednesday. We have sessions in the morning and afternoon, being run by volunteers, and its all completely free.
If you have kids who might be interested, you can sign up here. If you know anyone else who has kids who might be interested, forward this to them.
This year, Mobility Camp is coming to Sheffield on 30th September, and it promises to be another amazing day of discussion, debate, and taking action. Anyone with an interest in transport should be there, and so should you. You can get tickets here.
Unless you have been living under a rock for the last six months or so, one of the biggest projects currently in transport planning is the planned Mini Switzerland project in the Hope Valley. For those of you who have somehow missed it, Thomas Ableman - who came up with the idea and has worked with the Hope Valley Climate Action Group and many, many volunteers to make it work - describes it as so:
The project represents an opportunity for the East Midlands to demonstrate national leadership in rural mobility by hosting the UK’s first real-world demonstrator of fully integrated bus and rail services outside of a major city.
Unlike many transport schemes, Mini Switzerland does not depend on new railways or large-scale construction. Instead, it focuses on better coordination of enhanced services, supported by modest infrastructure improvements, clear information and simple, integrated ticketing.
I am naturally favourable towards the kinds of projects that back up a sky-high ambition with lots of good work by committed teams to make it happen, even if such projects ultimately do not succeed. In fact, it is my own failure that has taught me almost everything I know about operating integrated public transport services.
This is the story when, in one corner of Bedfordshire, we tried integration, and the result was a bus that carried virtually nobody. The Leighton Buzzard Station Travel Plan - one of the national pilot station travel plans and was award-winning to boot - contained a lot of ideas around how we would integrate the station into work being undertaken by the Cycle Demonstration Town, a new “Dash Direct” bus service to and from a development to the South of Leighton Buzzard, and personalised journey planning.
As we were delivering the station travel plan, we decided that we needed to trial a bus service from rural villages. As Leighton Buzzard is the last stop on fast commuter services into London Euston (taking as little as thirty-five minutes), it had a huge rural hinterland. We chose an obvious candidate village, and immediately started work to design the service. The bus in question was numbered the 27, and it operated a peak hour service between the village of Heath and Reach just to the north of Leighton Buzzard, and Leighton Buzzard railway station1.
Operated by the-then new bus company Grant Palmer, it ran twice a day in each direction. It was timed specifically to meet the busiest trains to and from London Euston - which we knew from spending sixteen hours a day for three days on station platforms counting the number of people getting on and off trains serving the station.
We even sorted the fares. The service accepted PlusBus, the add-on ticket that lets a rail passenger buy a day or more of local bus travel alongside their train ticket at a discount, so the money side of the interchange was about as close to seamless as the system at the time allowed.
On paper it was a model of integration. A dedicated service, timed to the trains, with a through fare already in place. What could go wrong?
It never carried more than about seven people a day. After a few years it was quietly withdrawn.
I have thought about the 27 a great deal since, because it tells me a lot about integration. We treat integration as a problem of ticketing and provision. Put on the bus, sort the fare, line up the times, and passengers will come. The 27 had all three, and they did not come. So the interesting question is not what we did wrong on the surface, but what lurked underneath the surface that scuppered our well-laid plans.
When a service does this badly you owe it to yourself to find out why, so we literally went out and asked people - both at the station and in the village. For some the timing simply did not work. A twice-daily service, however well aimed at the peak, suits only a narrow band of journeys, and plenty of villagers start or finish work at hours the 27 did not serve. For others the honest answer was that they preferred to drive, and a car on the drive next to a station with parking is a powerful thing to compete with.
But the answer that came up most often was that people did not trust it. Not that they thought the bus would fail to turn up necessarily (though it was mentioned), but that they could not see what happened when things went wrong.
The question people kept asking themselves was a simple one: what if the train was late? If you have driven to the station you shrug and walk to your car. If you are trying to catch the last bus of the evening and your train crawls into the station twenty minutes behind schedule, the bus has gone and you now have the task of trying to get a taxi, or calling home in the hope someone can give you a lift. One passenger told us they had missed the last bus for exactly that reason, because their train was late and the 27 could not wait. They never used the bus again after that.
In theory there was a safety net. The 150, the long-standing interurban route that threads through Heath and Reach between Aylesbury and Milton Keynes, offered a fallback. But there was a catch. Catching the 150 back then meant a ten-minute walk across town to the stop on Wing Road, in the wrong direction, at the end of a long day2. A tired commuter will not do that on a wet, cold Tuesday evening in December. So they did the rational thing. They took the car, or never started using the bus at all, because the downside of being caught out once outweighed the upside of a smooth journey most days. Those who did give it a try often bailed once it failed. The 27 was reliable on a good day and useless on a bad one, and people build their travel habits around the bad days.
The things we usually treat as the hard work of integration are actually the easy part, and on the 27 we did them all. Ticketing is largely solved. PlusBus is not perfect, but it exists and offers integrated ticketing. Provision is solved in the narrow sense that we can commission a bus whenever we find the money, which is what the station travel plan did. And timing is solved in the sense that any competent scheduler can build a timetable that meets a train. We did that too, down to the specific London services.
So if ticketing, provision and timing are all things we can deliver, and we delivered them, and the service still failed to earn trust, then the real barrier to integration lies somewhere we do not usually look.
Lets start by looking at the commercial structure of bus operators. Since the Transport Act 1985 deregulated local buses everywhere outside London, the default is a commercial market. An operator runs the routes it chooses, registers them, and keeps the fares. Where that leaves a gap the authority judges socially necessary, the authority can pay an operator to fill it through a tendered contract, so long as it does not directly compete with a commercial service. The 27 was one of those, a supported peak service bought to serve a policy goal rather than a commercial one.
The trouble is that a connection between a bus and a train sits in a gap neither the bus or train operator is paid to close. They are separate businesses with separate revenue. There is no mechanism by which the rail company compensates the bus company for holding a departure, and no shared pot that both draw from when a smooth interchange delivers a passenger who might otherwise have driven. In fact for the train operator, they get the passenger regardless, and may also keep more revenue through car parking.
When there is no commercial agreement to protect a connection, the connection is nobody’s job. It is a benefit that lands on the passenger and the wider network but not on the balance sheet of whoever bears the cost. As the Centre for Cities has argued, deregulation left public transport competing with itself, without the co-ordinated ticketing, pricing and timetabling that integration depends on. A held connection only happens when someone is responsible for the whole journey, and in our system usually no one is.
Then there is the punctuality regime. Bus punctuality in Great Britain is measured against a standard set by the Senior Traffic Commissioner. The rule, essentially unchanged since the practice direction of the mid-2000s, is a window of tolerance. A bus should depart its timing points no more than one minute early and no more than five minutes late, on at least ninety-five per cent of departures.
Missing that does not mean only a stern warning or disappointment from the Traffic Commissioner. Miss it consistently and the operator can be called to account, with financial penalties and, in the worst cases, action against its licence. In the worst cases, operators can have their licences revoked.
The challenge this poses is that the bus timetable does not account for the operational reality of other public transport systems. A driver who holds for eight minutes for passengers on a delayed London train has, in the eyes of the regime that judges them, simply run a late bus. The very behaviour that would build trust with rail passengers is the behaviour the compliance system discourages. We ask bus operators to integrate with the railway while measuring them against a clock that ignores the railway entirely.
It’s easy at this point to be critical of this regime, and many people I know have suggested that Traffic Commissioners should cut bus operators some slack for holding buses to meet trains. But there are very good reasons why this regime exists. The window also exists to stop early running, which strands passengers who arrive on time to find the bus gone, and the evidence from Transport Focus is that passengers hate early running even more than lateness.
Furthermore, even if we swept the punctuality problem aside, a second obstacle would remain. A bus that waits at the station is not sitting in isolation. It is usually part way along a route, with passengers already aboard and more waiting down the line. Hold it for a late train and you delay every one of them.
Let’s use the 150 bus as an example, and assume in a perfect world that, at the time, this hourly bus did what we planned the 27 to do. If it paused at the station to scoop up rail passengers whenever a train ran late, it would arrive late into Heath and Reach and everywhere beyond, to the frustration of the villagers already on board who simply want to get home. The driver is being asked to trade a certain, immediate harm to the people sat on the bus for an uncertain benefit to a handful of strangers who may or may not be on the delayed train. That is a hard trade to make on instinct, and the deregulated route structure, in which a single bus serves many masters along its length, makes it harder still.
A further barrier seems trivial, yet in our experience played a bigger role than we expected. Even a driver who wanted to wait, and was willing to absorb the risk, usually has no way of knowing whether the train is late. There is no feed of live rail running information into the cab, and using the mobile at the wheel is an obvious no-no. To see if the trains were running late, the bus driver would literally have had to lock up the bus, walk into the station, and check the departure board themselves.
This is doubly frustrating because what we have created in terms of data infrastructure runs the opposite way. The Bus Open Data Service means a bus broadcasts its own position and predicted times to journey planners, to bustimes.org, and to the Traffic Commissioner’s punctuality monitoring. We have built an entire system to tell the world where the buses are. We have not built the far simpler thing that would let a driver make an informed decision about a connection, a signal that says the 18:42 from Euston is running nine minutes late, hold if you can. Without it, a willing driver is guessing, and guessing wrong means either a pointless delay to onboard passengers or a missed connection anyway. So the rational driver does not chance it.
Underneath all of this sits a commercial logic that leaves no room for waiting even in principle. Bus operations are planned around the fewest vehicles and tightest driver duties that will deliver the timetable, because vehicles and drivers are the two biggest costs in the business. Recovery time, the slack that lets a service absorb a delay and still start its next trip on time, is minimised because it looks like idle resource. A schedule with enough slack to hold routinely for late trains is carrying spare capacity, and that is the first thing a cost-conscious operator engineers out.
In our case, the bus that operated the 27 was frequently used for other routes during the day, before coming back in the evening to pick up commuters. It could not afford to run late for the second service of the morning, otherwise its whole service pattern for the day would be affected.
Finally, there is the fragility of the thing we commissioned. A peak-only bus running twice a day is, by design, a marginal service. It carries a high subsidy per passenger, it is the first candidate for the axe when budgets tighten, and it can never generate the habitual, turn-up-and-go ridership a frequent all-day service builds almost automatically.
Trust in a bus is largely a function of frequency. If there is another one in ten minutes, a missed connection is an annoyance. If the next one is tomorrow morning, it is a catastrophe, and people plan their whole travel choice around avoiding it. We ask a thin service to prove demand it can never build because it is too thin to be trusted.
Lay these barriers side by side and a pattern emerges. Commercial structure, punctuality rules, route interdependence, information gaps, cost pressure and marginal frequency all push the same way. Each, on its own, makes it a little more rational for an operator not to protect a connection. Together they make the trustworthy behaviour almost impossible, not because anyone is being obstructive, but because the incentives point the other way.
And notice where the failure concentrates. The late running train unravels the whole service. On a good day the 27 worked beautifully. Its problem was that integration which only works on a good day is not really integration. It is two timetables that happen to line up. The value of a genuine connection lies almost entirely in what happens when things go wrong, which is exactly what our system handles worst.
The Mini Switzerland project is seeking to apply to the Hope Valley the lessons learned in the country contained within its name. The contrast is not really about Swiss efficiency or Swiss money, though there is plenty of both. It is about what the system is designed to optimise.
Switzerland built its network around the connection rather than the vehicle. The organising idea is the Taktfahrplan, the clock-face timetable introduced nationally in 1982, in which services run at regular intervals and are timed to meet each other at designated nodes. The Swiss call these nodes Knoten, and stations like Olten, Bern and Zurich are engineered so that trains from several directions arrive together, exchange passengers within a few minutes, and depart together. The whole timetable pulses, and buses are woven into the same rhythm, so the Postbus in the valley is built to meet the train at the station, not by luck but by design.
The philosophy behind this inverts our instinct. Under the Rail 2000 programme the Swiss adopted the principle that trains should run as fast as necessary, not as fast as possible. Rather than chasing raw speed and hoping the connections fell out of it, they fixed the connection times first and then invested in the track and signalling needed to hit them.
But the part that matters most for the 27 is not the timetable at all. It is the governance and the money underneath it. Swiss regional transport is organised through tariff associations, with public authorities specifying and funding an integrated network, so revenue is pooled across operators and modes rather than fought over. When the money is shared, an operator that holds for a connection is not handing a gift to a rival at its own expense. The benefit and the cost sit inside the same system, so waiting becomes affordable because the incentive to wait is built into who gets paid. One body owns the journey, so the connection is somebody’s job.
It should be noted that even in this world class system, the Swiss do not hold buses indefinitely. PostAuto itself notes that because the timetable is so tightly interlocked there is only limited scope to wait, and delays are handled through national compensation rules rather than bus drivers waiting for twenty minutes at stations. The lesson is not that Swiss buses wait forever. It is that the system is punctual and co-ordinated enough that holds are rarely needed and, when they are, short.
That reliability is bought with sustained investment and a settled governance structure Britain does not have, and the honest caveat is that our own evidence base on integration of this kind is thin. Most of the good practice we cite is imported, from Switzerland, Germany and the Netherlands, and it needs adapting rather than copying.
The system incentivises connections and accessibility. Not speed. And the governance is built around that goal.
We reach, again and again, for the parts of integration that are legible and deliverable. We sort the ticketing, we commission the service, we time it to the trains, and we present that as integration being done, because those things photograph well in a travel plan. What we reach for far less often is the incentive structure underneath, the commercial agreements, the punctuality rules, the information flows and the funding models that actually determine whether an operator behaves in a way that earns a passenger’s trust.
And trust is the whole game. If the train is running late and you know, with confidence, that the bus will wait, that is a categorically different proposition from the one we offered in Heath and Reach. Under our arrangements the bus cannot promise to wait, because every incentive it faces tells it not to. We built the 27 to generate trust inside a system that gives operators every reason not to provide it.
None of the fixes are exotic, they are just tricky. A contractual connection guarantee with a funded allowance to hold. A live rail feed into the cab so a driver can make an informed choice. A punctuality standard that recognises a protected connection rather than punishing it. Revenue pooling so that waiting does not bankrupt the operator that waits. The franchising and Enhanced Partnership powers in the Bus Services Act 2017, used to put one body in charge of the whole journey. Each is a change to an incentive, not to a timetable or a ticket.
The 27 was not a failure of ticketing, and it was not a failure of scheduling. Those we got right. It was a service asked to do the one thing our system is built to prevent, and to be trusted when things go wrong. Until we are willing to work on the incentives and governance rather than the basics, we will keep commissioning careful, well-timed, properly ticketed little buses and watching them carry five people a day until the money runs out. This village bus taught me that the timetable was never the hard part. The hard part was everything we did not put on the timetable.
While I love writing this, and my writing will always remain free to view, if you are able to donate whatever little you can to help out it is much appreciated. I do have bills to pay after all! You can do that right here.
👩🎓 Further Reading
The clever clogs at our universities, government departments, and other clever people have published the following excellent research. Where you are unable to access the research, email the author – they may give you a copy of the research paper for free.
Creating bus timetables with maximal synchronization
TL:DR - The authors turned the intuitive goal of making buses meet into a formal optimisation problem, maximising the number of simultaneous arrivals at connection nodes and solving it as mixed integer linear programming.
TL:DR - The penalty a passenger attaches to a transfer depends on its quality, not merely its existence, and that planning practice routinely lacks a clear behavioural framework linking the design of stops and interchanges to how people actually decide to travel.
TL:DR - Puts hard numbers on the psychological cost of transfers on the London Underground, finding that a large share of the burden is not walking or waiting time but the penalty passengers assign to the act of transferring itself.
Users’ willingness to ride an integrated public-transport service: a literature review
TL:DR - Both research and practice have over-concentrated on the operational mechanics of transfers and under-studied the psychological and policy factors that actually shape whether people will use an integrated service at all.
TL:DR - Time and cost savings on their own do not reliably persuade people onto routes that involve a transfer, because reliability and confidence carry heavy weight in the decision.
😀 Positive News
Your weekly dose of positivity in a world where bad news dominates. Enjoy.
Greater Manchester’s franchised buses keep growing
Bee Network bus patronage rising over the rolling twelve months from 168.9 million journeys in January 2025 to 176.3 million in January 2026, with passenger numbers up eight per cent year on year in the first tranche of franchised areas and ten per cent in the second. Metrolink recorded its busiest January on record at 3.7 million journeys, and the network now runs affordable capped contactless ticketing across bus and tram.
Contactless pay as you go spreads across the South East
The Rail Delivery Group confirmed that tap in, tap out contactless ticketing has gone live at thirty further stations, taking the total to fifty three, with a further twenty including Stansted and Southend airports to follow this summer 2026. More than 5.6 million contactless journeys have been made since the rollout began in June 2024.
Liverpool City Region brings back franchised buses after 40 years
Merseytravel named the operators for its first franchised services in four decades, with invitations to tender for the second tranche covering Knowsley, Liverpool and Sefton commencing in February 2026. The first phase begins in St Helens and Wirral in autumn 2026, rolling out region wide by the end of 2027.
The West Midlands moves to franchising with an electric bus rapid transit backbone
Transport for West Midlands issued its first tranche invitation to tender in May 2026, with reliability, better integration and more affordable fares set as the primary goals. The plan is anchored by Sprint, a bus rapid transit scheme for the A34 and A45 corridor running 24 tram like battery electric articulated buses with dedicated charging infrastructure.
Germany's national ticket holds firm through a price rise
The Deutschlandticket, the single monthly pass valid on almost all local and regional buses, trams, S-Bahn, U-Bahn and regional trains across Germany, retained more than 13 million active subscribers after its price rose to 63EUR a month in January 2026. Estimates from mid 2025 suggested that around a fifth of trips made on the ticket were net additions to public transport rather than journeys shifted from existing tickets.
💻 Hard Work
As I mentioned last week, I probably won’t write this section for a couple of weeks due to the reason given in my introduction. So if you really want to hear what I have been up to which isn’t meetings, Teams calls, and riding up and down railways, you are going to have to wait. Sorry.
🎶 Musical Finale
Sorry to say that Bachelor Girl’s song Buses and Trains is not about buses and trains. It’s about boring things like relationships. Who’d sing about those kind of things?
🪧 Notices
This section is intended to notify you of important things that may or may not annoy you, depending on your view of the world. But these are things I feel it is worthwhile pointing out.
Regarding Artificial Intelligence. For my newsletter, I use AI like a research assistant. I get the AI (in my case Claude Opus 4.8) to find for me well-cited and evidenced research in the subject matter, focussed on academic journals with a high impact factor. I follow this up with my own research, and I check each source myself. Where I have the time to, I contact domain experts or seek out their published works to validate the findings. For this particular article, I did get AI to write a first draft of some paragraphs, to give some ideas for structuring the article, which I subsequently edited into my own words. Otherwise, the words written here are my own, spelling mistakes and all.
Regarding my expertise. I am a transport planner, so my realm is public policy, how people and things move, and their impacts. While I research everything in a manner I consider robust, there may be some instances where my lack of domain expertise fails me. That’s on me, but that’s all part of learning. I always encourage any reader to do their own research on the subject matter.
✅ Sign Up
If you have been sent this newsletter, please consider signing up. Its free to do so, and I don’t spam your inbox with random rubbish - just my own words.
Fun fact: Leighton Buzzard station is not actually in Leighton Buzzard, it is in Linslade. Which, up until the local government reorganisation of 1974, was an entirely separate parish. The whole parish is now called Leighton-Linslade, with the town of Leighton Buzzard and the village of Linslade now being covered by a Town Council. I’m sure it made sense at the time.
In a wonderful irony, the 150 now stops in the forecourt of the station, after Arriva told us for years that this was not possible in the timetable they had.
In a former life and a different century, we organised a more successful bus service from TRING station - a short distance south of Leighton Buzzard. The existing Tring Town Service (it’s a large village, really!) was extended to the station of the same name, a couple of miles outside the town. This extended service prospered because it was an extension of the town service, which provided a base of existing usage and operated along many roads in the village providing good accessibility for intending rail passengers. It was scheduled to meet the trains and operated twice an hour, 14 hours a day. I don't think the fares were integrated, but bus season tickets were priced to sell.
I don't know why anyone else didn't think of it, but we tapped into the station departure screen system and installed an extra screen adjacent to the bus stop, where the driver as well as intending passengers could see it. We established clear protocols for maximum waiting times in the event of delayed trains and the driver could easily work out if it made sense to wait for the train. Advice that the bus would wait for delayed trains was included in the printed timetables and roadside publicity, so train passengers were re-assured and passengers along the route knew that the bus might operate later in specific situations.
The timetable had recovery time built in, so timekeeping was maintained. As the service was operated as a cost contract, the operator followed the protocol and everyone was happy, especially the passengers. Over the years the service was salami sliced to achieve cost savings and the attractiveness reduced.
One issue we couldn't solve at the time was that passengers further along the route did not know if the bus was 'delayed' because the driver had waited for a late train. Technology now has an answer to this and RTI screens at bus stops and apps can now also keep non-train passengers informed of the status of the bus. All it takes is for the local authority to be passenger focused and some bus drivers who are still capable of a little mental arithmetic, though a modern system with a schedule adherence function could take over that task too.
I do think the impact of not just frequency but reliability has much more of an impact on passenger behaviour than is generally discussed, as you mentioned.
This is anecdotal, but my (rural) bus route used to run 4 times per direction per day to and from the city on Sundays, but constantly had driver absence issues that meant that service got degraded. The last bus back was at around 17:15, but I would ALWAYS plan my day around getting the previous 15:15 service instead, so that if that service wasn't running I'd still be able to get the last bus two hours later, or find alternative arrangements if necessary.
A few years ago, that service was bumped from 4 to 6 buses per direction per day, with a new later 18:15 service being the final journey. I still mostly plan for the 15:15 because it's Sunday and not much happens after 16:00, but I also not-infrequently plan around the 17:15 instead, knowing that I have a fail-safe in the 18:15. It's not a 100% reliable strategy (the next bus could be cancelled too of course), but it hasn't failed me thus far, and that safety net is a POWERFUL anchor cognitively.
A similar anchor in the opposite direction: my most common rail journey, I usually buy on split Advance tickets, for a regional leg and then a long-distance leg. The regional leg is frequent enough and easy enough to research that I will look up how the trains are running in the morning, and potentially make the decision to abandon the Advance ticket and travel earlier (buying an Off-Peak or Anytime single) to guarantee making my booked long-distance service.
Now I am an unusually anxious person, to some extent, and I'm also an over planner, but I have seen others who aren't quite as neurotic as I am who apply some of the same principles to lesser extents. I'm also probably at the extreme end of price-sensitivity, being unwaged and travelling almost exclusively for leisure and family reasons.