Why did Bedford lose its peak express services?

This article looks at one of the three major problems that have hit, or will hit, railway users in Bedford. In reverse chronological order of when they hit, they are: the loss of direct intercity trains to destinations north of Kettering; the failed delivery of the May 2018 timetable changes; and the announcement at the end of last year that as part of the May changes, Bedford would lose its peak time express services in the direction of the peak (towards London in the morning, from London in the evening). Although the three things can be defined separately, in fact they are closely interconnected: this article tries to unravel the reasons for the peak changes, and the next one will look at the future of northbound services and other things to do with the next East Midlands franchise.

But first, some thoughts for context. The announcement of the changes in December was highly unusual: it was made at short notice, with no consultation – indeed, the consultation on the next East Midlands franchise had recently been run, and there was no mention of any changes like this. Is there any precedent for such a sudden and major withdrawal of services from any major conurbation in the history of railways in Britain? So, whatever the reasons for this highly unusual announcement, they too are probably unusual.

It’s also got to be observed that this as classic an example as one can get of faceless bureaucracies treating ordinary people with contempt. Behaviour of this sort is exactly the sort of thing that supposedly unaccountable and unresponsive nationalised industries used to get accused of – but today’s railways are the product of the 1980s’ and 90s’ drive towards privatisation and ‘new public management’ that were supposed to stop public services dumping on people in this way. Hasn’t worked, has it? It’s also an illustration of how passengers are more or less completely cut out from the running of the railways (on which more another time).

Nor are Bedford’s travellers even clearly the worst hit: people travelling into Bedford from the north in peak times, for whom the East Midlands expresses were the only option, don’t have any trains at all now and are stuck on a much slower rail replacement bus service (ironically this includes a chunk of Thameslink’s drivers who sign in at Bedford…). The faster Thameslink services that have been scheduled to make up, in theory, for the lack of expresses, have only been created by removing stops at Flitwick, Harlington, Leagrave and Harpenden, which have suffered an overall reduction in services as a result – Harpenden commuters in particular are furious about it. Now, all timetable changes create winners and losers, and it’s the anguished screams of the losers that get heard; but it’s really noticeable just how many losers and how few winners the May timetable would apparently have created on the Midland Main Line even if it had been delivered properly.

This post is split into two parts. The first looks at why the peak changes are required from a practical, operational point of view: what does the railway gain by cancelling stops at Bedford (and Luton)? The second looks at why the changes happened from a decision-making point of view: whatever the operational merits of the changes, what was the process that led to such short-term, chaotic decisions? There’s a third part to this, which is the question of what the future will hold for express trains calling at Bedford – but that is really part of another article, looking at the plans for the next East Midlands franchise.

Part One: service patterns and logistics – why the changes are needed
In one sense, the peak changes happened because something went wrong. To understand what that something was (or range of somethings were), we first need to look at what should have happened. In January 2013, the plan was for the Midland Main Line to be electrified from Bedford to Corby in 2017, Kettering to Derby and Nottingham in 2019 and from Derby up to Sheffield in 2020. The collapse of these plans also triggered, ultimately, the ongoing debacle about rolling stock on the route, as previously explored.

The High Level Output Specification for Control Period 5 (CP5 – the planning period for Network Rail during which the electrification should have taken place) outlines several key changes that should have been delivered as part of the electrification works. At present, track layout and signalling configurations, plus restrictions imposed by other routes using the same sections of track (eg Thameslink between Bedford and St Pancras) limit the East Midlands service to five trains per hour (tph) into St Pancras, and five in the other direction – typically that involves two Nottingham services, two Sheffield services (via Derby), and a Corby service. The HLOS for CP5 aims for a sixth path, going to Corby, giving both an ‘enhanced Corby-Kettering-London service’ and ‘accelerated Sheffield-Derby-London and Nottingham-London services’. Implicit in this is the idea, all the way back in 2012, of cutting out stops on the long-distance expresses and instead having Bedford and Luton served by Corby trains, as was eventually proposed in the 2017 consultation on the new East Midlands franchise.

Illustration in the Atkins report of the proposed service pattern, splitting Bedford and Luton off into the Corby services and sending the long distance expresses through non-stop. Click for larger view.

Another particularly illuminating document here is this economic case prepared in 2016 for the Department for Transport by the rail engineering and consultancy company Atkins, while full electrification was still planned (after the ‘Hendy pause’ of 2015). This makes the split in services explicit, and emphasises that it is essential to realising the benefits of the scheme – ie the benefits that are necessary in order for it to be good value for money. This was forecast to generate higher passenger numbers, and be key to the ‘very high’ value for money offered by the scheme; the analysis suggests that these higher passenger numbers will more than outweigh any losses from the end of direct services between, say, Luton Airport Parkway and Leicester, Nottingham, Derby etc. This, then, is the reason for the proposed split of services that will mean the long-distance services no longer stop at Bedford, which was clearly being planned well before the publication of the 2017 franchise consultation, if you knew where to look.

So, with the line to Corby electrified and the middle-distance services segregated from the long-distance expresses, the East Midlands franchise timetable should then have dovetailed, in May 2018, with the expansion of Thameslink services. Prior to May this year these ran at 15tph each way between St Pancras and stations to Bedford in peak hours, but were due to go up to 16tph on the Midland Main Line (as part of the push to 24tph through the central London ‘core’). So, we should have had 16tph on Thameslink plus 6tph on the East Midlands franchise, two of which would have run to Corby.

It’s also important to think about the sequence of those services: you can’t have a fast service immediately behind a slower one, as obviously the slower one will hold the faster one up. So there should be two fast Thameslink services (with relatively few stops) ‘flighted’ together, overtaking two slow Thameslink services, and then two East Midlands Trains services ‘flighted’ together.

But of course that’s not how it happened. The electrification to Corby is still happening, but much delayed. Currently the Midland Main Line has only the capacity it previously did. The Thameslink programme, however, has been to a large extent completed. This is where the introduction of the notorious May 2018 timetable and the cancellation of the electrification scheme combine poisonously: the introduction of the new Thameslink services has obliged East Midlands Trains to try and fit in their new service pattern around the Thameslink one – except without the electrification and associated works, they can’t do it as they’d envisaged. So now they’re trying to run the ‘improved’ service pattern, minus the additional train per hour, and much less efficiently because of constraints imposed by Thameslink.

A few examples will give you an idea of how the new timetable is much less efficient for East Midlands Trains than the old one. For instance, a train arriving at Corby from St Pancras would previously have spent seven minutes there before turning back towards London. Now it’s there for 45 minutes (though it would be less if Corby had its 2 tph), tying up an entire train for an extra chunk of time. As was explored in the rolling stock article, the High Speed Trains (HSTs) on the route have worse acceleration than the more modern Meridians – this means they have to be used on specific services with more generous timings to allow for this, and the net result is that layovers in Nottingham of an hour under the old timetable have turned into an hour and ten minutes under the new one. Hence East Midlands Trains have taken on three extra HSTs – with the new and less efficient timetable, they have needed more rolling stock just to operate the old level of service.

These issues have combined to force the lack of peak calls at Bedford and Luton – although curiously, for different reasons in the morning and evening peaks. Starting with Bedford and the morning peak, the key issue is that the East Midlands services can’t be made to fit in with the Thameslink ones. The track layout of the station provides a dedicated platform for northbound expresses, which no Thameslink trains ever use (indeed, they can’t – it’s not electrified). Southbound, however, is a different matter: southbound expresses can use any one of three platforms, but all of them are shared with Thameslink services. Delays to Thameslink (a significant danger, we now know) therefore risks the express getting stuck outside Bedford station, waiting for a platform to become free. However, there is an extra line that goes through Bedford and does not have a platform next to it at all. It can be used for trains in either direction – very obviously, if the southbound trains are able to use that track, non-stop, they won’t get stuck waiting for a platform to come free, and therefore won’t be at risk of being delayed at Bedford. So that’s what is now being done.

It’s not just the track layout at Bedford station itself that’s a problem, however. There are two sets of tracks running to London – fast and slow, one of each in both directions. If a fast Thameslink service gets delayed under the new timetable, switching it to the slow lines to get it out of the way of an East Midlands express is harder than it was under the old timetable – with more intensive Thameslink services (in theory), there are fewer gaps on the slow lines to slot a delayed fast service into. This means that once an East Midlands express has been held up at or south of Bedford, it can’t make the time up again as was previously possible, and is guaranteed to get into London late. This gives a double whammy of reasons why southbound East Midlands services are now being sent non-stop through Bedford in the morning peak.

In the evening, the problem is different, and related to those inefficiencies in how the new timetable uses its rolling stock. The four peak expresses that used to call at Luton and Bedford in the evening used to be made up of two five-coach Meridians coupled together, giving ten coaches in total. Under the new timetable, the equivalent services are short-formed (though I’m unsure whether this means one five-coach Meridian or one of the six-coach HSTs – probably a single Meridian). They therefore can’t accommodate the number of passengers that would catch these services to Bedford and Luton. Luton Airport Parkway, however, is still being served – presumably those trains (not the same services) are long enough to accommodate some long-distance commuters (though quite how many people would try to get to Luton Airport Parkway on an express and then switch to Thameslink to get to Luton is unclear – certainly the same switch doesn’t get you up to Bedford any quicker).

So, from an operational point of view there are strong reasons for the changes being made, albeit that they are also technical and a bit obscure unless you look at them hard. But why were those timetable limitations imposed in the first place, and how did the decision-making end up being so chaotic that the only way to sort out the problems was a very late imposition of changes that seriously degraded many people’s rail services?

Part Two: the decision-making
Amid all the recriminations following this May’s timetable fiasco, Rob Warnes, the Director of Performance and Planning and Management at Northern Rail, made an important point when giving evidence to the Commons Transport Select Committee. Northern’s slower, local services are ‘last on the graph’ when timetables are planned: so, when TransPennine Express (which operates faster services between the North West and, variously, Yorkshire and Scotland) reconfigured the pattern of its faster, longer-distance trains on the same stretches of line, Northern had to fit its services in around them. Which stands to reason: fitting faster services in around slower services will just lead to the faster services getting held up too often, so you have to do it the other way round.

But on the Midland Main Line, this sensible way of doing things was up-ended: East Midlands Trains were required to fit in around the expanded Thameslink service – or at least, what was expected to be an expanded Thameslink service. As one commentator has put it, the Thameslink tail is wagging the East Midlands dog. We’ll see now how that unfolded over time, leading up to the withdrawal of the peak services from Bedford.

Let’s go back to spring 2016. According to London Reconnections, by that time it was already apparent to a good number of people in the industry that Thameslink’s timetable change in May 2018 would not be feasible – largely because the new rolling stock for the Thameslink programme was being delivered late, and it wasn’t going to be possible to make up the delays. But at this point the actual planning for the operation of Thameslink hadn’t yet started in earnest.

The House of Commons Public Accounts Committee issued an update to its previous report on Thameslink in February this year – before the timetable unravelled. It found that the Department for Transport and Network Rail took too long to start planning how the newly expanded Thameslink would operate. The detailed planning started only in late 2016, when an Industry Readiness Board was established: this was the first time that the full range of parties in the wider rail industry were brought together. The Committee also found that the Thameslink programme’s governance arrangements were also inadequate for providing clear responsibilities in terms of who would make what decisions – a problem only resolved by a review of governance in mid-2017. Both elements – the Board and the clarified governance arrangements – needed to be in place much sooner.

What the industry had to work with when it eventually got its act together was the 2016 Atkins report that updated the business case for Midland Main Line electrification and associated upgrades, and also provided a draft timetable based on initial planning from Thameslink (ie done by Thameslink in isolation, without any reference to the wider railway). The Atkins timetable attempted to fit EMT services in around the Thameslink one, but featured clashes between trains, particularly in the evening peak when Thameslink trains had to make crossing moves from the slow to fast lines at Harpenden and Carlton Road (near West Hampstead). These clashes were never resolved by Atkins. The Industry Readiness Board picked up the task only in spring 2017 and, crucially, mandated East Midlands Trains to resolve the issue as much as possible, keeping the Thameslink service intact. The crucial factor here may be that the Readiness Board was mandated to oversee the introduction of the expanded service on Thameslink, and therefore prioritised it above other operators.

However, there is no clear and direct answer to the question of who made the decision to withdraw peak services, beyond that it was East Midlands Trains – in as much as they were EMT services in the first place, to the call to ditch them had to be made by EMT. In a Westminster Hall debate on the peak changes in Parliament on January 23rd, Gavin Shuker MP recounted a meeting he and Alisair Burt MP had had with Thameslink and EMT, which failed to arrive at an answer to the question of who made the decision.

Did the Industry Readiness Board effectively pull the trigger on the peak services at Bedford, or did it simply not understand the likely consequences of its efforts to get Thameslink up and running, and how they interplayed with other factors such as the delivery setbacks to the Midland Main Line upgrade? It certainly appears that neither the Department for Transport nor Network Rail had the capacity, and still less the mandate, to monitor, foresee or prevent the changes. Ultimately, it feels very much as though Bedford’s commuters have become collateral damage from a set of processes that have little or no cognisance of them, and that the fundamental structure of the railways is set up so as to be incapable of identifying threats to passengers’ services and to acting on their behalf. Some remarkable cack-handedness at multiple stages, not least the late start to planning by DfT, ultimately left withdrawing the peak services as the only available bodge.

But there’s some unfinished business here. The proposal that Bedford should be served only by St Pancras-Corby services (in terms of the East Midlands franchise) was bound up with the decision-making that led to the loss of the peak expresses, and the ‘Corby split’ plan seems still to be integral to how they will be reintroduced. The future of these services will be revisited in a subsequent article.

Back when you could get a peak express from Bedford, in two senses. Little visual joke for you there. Adapted from a photo by Mark Harrington under Creative Commons licence BY-SA 2.0, and available for reuse under the same.

3 thoughts on “Why did Bedford lose its peak express services?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.