Meltdown Monday: How Southern’s Problems Run Deeper Than Disputes


For those living and working in London and the South East it has become almost impossible to be unaffected, either directly or indirectly, by the trouble-hit GTR franchise. Understandably there is anger and frustration at what is going on.

The problems on Southern that have led to a controversial emergency timetable being introduced are often portrayed in very simple terms. At its simplest, it would appear to be a petty dispute over who should shut the doors on the train. This simply does not stand up to scrutiny, however, as a complete explanation when considering all that is going on. A dispute about trains involving the guard’s function would not, for example, explain why there have been substantial service withdrawals in inner London on routes where the trains are driver only operated. Nor would it explain the numerous daily cancellations into stations like Moorgate (Great Northern) which is entirely served by driver only operated trains.

Pebbles and avalanches

Modern life thrives on the twin concepts of efficiency and optimisation. Making the best use of what we have is a laudable aim – whether it is with regards to energy supply, road use, assembly line production or getting the most out of our railways. There is, however, a big downside to this desire to avoid wasteful or inefficient practice – when something goes wrong, what should be a minor problem may become a major one. Worse still, when two or three things go wrong simultaneously it can be extremely difficult to restore order and normality.

People have warned for years that London’s transport system will start to collapse due to the sheer number of people using it. That is unlikely and gentle degradation is a more likely outcome. What consistently seems to get overlooked, however, is the possibility that two or three problems, relatively small and insignificant in themselves, can come together to produce a situation that is hard to unravel and even more difficult to solve. The miserable experience of passengers at London Bridge last year was an example of this. Three or four factors, any of which, on their own, weren’t too serious, combined to produce a toxic situation that involved much activity and protests from passengers, disgruntled user groups, MPs and other politicians.

When it comes to stability and reliability an increase in passenger numbers clearly does not help. In particular, a sudden rise is in passenger numbers, as encountered last year on c2c, is something that is less easy to deal with. It also means that any sudden change in one place (e.g. temporarily removing a service) can, if not well handled, quickly cause an unwelcome domino effect. Here, we take a look at seven problems that are, or may be, affecting Southern Railway. They are:

  • Driver Shortage
  • Rolling Stock issues
  • London Bridge
  • Ongoing Signalling Issues
  • Reactionary delay
  • Dwell time
  • Guards dispute

The first key decision is made

A favourite question amongst historians is “when did World War 2 start?” The question can be answered in so many ways because it is tricky to establish at what point war became inevitable. Not only that, dates and times are seen differently by different people. Even the time of the passing of Chamberlain’s ultimatum has to be put in context (it was actually at noon, Berlin time). In a similar way we have to look at current events south of the Thames in a wider context than one might think.

We could go back to privatisation and look at all the relevant issues that it created, including delaying the Thameslink Programme, but perhaps the first significant event was in April 2008. It was then that the Secretary of State for Transport, Ruth Kelly, approved the draft Thameslink Rolling Stock specification that was finalised later than year. This specified provision for Driver Only Operation (DOO). Wittingly, or unwittingly, the first seed had been sown that was to grow into an industrial dispute.

At the start of the franchise

It is probably best next to advance to the awarding of the Great Northern, Southern and Thameslink Railway franchise to Govia Thameslink Railway. Govia took over Thameslink and Great Northern in September 2014 just months before major changes at London Bridge. Southern Railway became part of this particular franchise in July 2015 but, as this was already being run by Govia, this was more a case of a change of franchise conditions rather than a change of management.

Prior to the problems at London Bridge in January 2015, Southern was generally regarded as a well run railway – one run by much the same people as it is now. This in itself should make it clear that the “Southern Railway are incompetent” point of view is, at best, simplistic and does not explain either the current problems or why it seemed to be operating quite well for many years up to 2015.

Where have all the drivers gone?

A recurring theme in this article will be the shortage of drivers – not guards. From a passenger’s perspective, the rot started when GTR took over Great Northern and Thameslink. Not highly publicised at the time was the fact that when GTR took over from First Capital Connect they were shocked to discover on day one that they were considerably short of their expected number of drivers. In evidence to a recent Select Committee, Dyan Crowther, Govia Chief Operating Oficer, spoke of having 607 when expecting over 650.

At the time, the initial shortage of drivers was given little publicity, but more recently GTR have not been so reticent about this. It does pose a lot of awkward questions though. How was First Capital Connect allowed to get away with this? Wasn’t there a basic franchise requirement for the previous franchisee to hand over the franchise in good order? Why didn’t GTR know about the problem in advance? Did they not ask the DfT? Did the DfT themselves not know?

Alarm bells ringing

It would surely have been a dejected Charles Horton, Chief Executive of GTR, who would have quickly grasped the reality of the situation on Day 1. He had far too few drivers. They were also coming up to the Christmas period when drivers are known to be more reluctant to work overtime. The franchise commitment required additional services to be run such as full weekend working to Moorgate. On top of this he must also have known that he was going to have to take drivers out of productive working for retraining on class 387 and subsequently Thameslink class 700 rolling stock. If that were not bad enough, there was a major change to the track layout due at London Bridge after Christmas and so drivers would have to be taken out of service to be briefed and trained on the changes there as well. For whilst the Thameslink core may have been closed, Thameslink trains still ran to London Bridge and terminated there.

He may have thought things could not get any worse. Unfortunately they did – a recurrent theme here in our story.

Sunday Working

One of the issues that still has not been tackled properly on the railways is the archaic practice of Sunday Rest Day working. Basically on large parts of the railway, still, Sunday is not a rostered working day for train crew, and management is reliant on people working rest days to provide a service.

Sunday Rest Day Working was a staple part of the old British Rail. The (generally male) train drivers were often happy to work the extra day for a day’s pay at overtime rates. It saved management having to employ a greater number of drivers and so there was a benefit from the savings possible with a smaller workforce.

In the past the service on Sundays was relatively sparse. The Sunday turns got worked either because of a local agreement not to fill up the vacancies in return for the union guaranteeing to fully cover Sunday working or simply because drivers didn’t want to risk not being selected by themselves being too selective as to whether to work it or not. Besides, Sunday was generally a day of rest with not much going on so one might as well work it and earn extra money. It was to the frustration of some commuters that there might be a train shortage during the week, but there was nearly always a full service on Sundays when few people travelled.

The first problem with relying on Sunday Rest Day Working nowadays is that on many routes the Sunday service is as intensive as the off-peak Monday-Friday service, even if not so many people use it and the trains themselves are shorter. So there are more duties to fill.

Probably more critical is that the social makeup and characteristics of train drivers and the world they (and it is they now, not ‘he’) live in has changed.

It is clear that Sunday Rest Day Working is an anachronism in the 21st century and in many train operating companies it has been abolished with Sunday now part of the working week. The trouble is that if the drivers don’t want to work Sundays as part of the rostered schedules (or they just don’t want to work on Sundays) then it is not easy to change conditions of service. Furthermore, if a franchise is only seven years long there is very little incentive for the train company to push for a change if it is liable to be acrimonious. It could well be argued that to abolish this practice it is necessary for the DfT to take the lead and require it as a condition of the franchise.

Even in the unlikely event of a train operating company having a full establishment strength of drivers, this is unlikely to be enough to run the full timetable. The figure is probably a historical one and rarely takes full account of time needed for training, annual leave (which peaks in the summer months) or an unexpected level of sickness higher than average.

Sickness can be a particular problem being unpredictable, to some extent. It is no use taking the average level of sickness and it could be considered prudent to allow for enough drivers for a full service to be possible for, say, 95% of the time – at least. With trains being such a vital part of daily life and penalties in place for not running a full service, it ought to be the case that it requires a substantial rise above the usual average (non peak) level before trains have to be cancelled.

There is also the issue of staff turnover. In his evidence to the Parliamentary Select Committee Charles Horton talked of an unexpectedly high level of 5.6% from a historical figure of 3.4% – presumably this is annual turnover. To a layman, even 5% seems incredibly optimistic in the modern day. Drivers retire, fail medicals, pass signals at danger and so either need further retraining or need to be permanently removed from train driving. Although not technically classified as turnover, some drivers need maternity leave and past data is probably not a reliable indication for the future provision required for this.

Lots of driver training in progress

Charles Horton emphasised that it takes 14 months to train a driver and claimed in his testimony that Southern have embarked on the biggest driver training programme of any Train Operating Company (TOC) in history. We would suggest that may not quite be true (MTR’s driver programme of driver training for Crossrail might be a valid contender) but it has to be said that it does appear from the relevant page on the Govia website that an enormous effort is being put into training new drivers – at least north of the river. Of course, a driver training programme in itself makes the short term situation worse as existing drivers are put in the classroom to become instructors for the next generation.

One still wonders what difference it would have made to the service on offer these days if Govia had started to prepare to train more drivers in May 2014 when they were awarded the franchise. You would still have had over a year of reduced driver availability, but at least training could have started in September 2014 when Govia started operations. This would have meant that by Chrismas 2015, when cancellations really started to get bad, Govia would have had more drivers available. By now they could have been up to full strength – at least on Thameslink and Great Northern services.

Going south

Meanwhile, on the Southern website we don’t seem to be seeing anything like the same level of driver training. It is dated December 2015. The latest (undated) update given on the website states that a further 78 are in training, which may well still be inadequate to cater for future needs. In evidence to the select committee Charles Horton quoted 82 drivers in training. A driver shortage on Southern Railway is harder to understand. After all Govia held the previous franchise and must have been aware of driver numbers.

Cancellation of services on Driver Only Operation routes

For many, the shock of the Southern emergency timetable was the cancellation of many services run by DOO trains. Indeed the first shock was that there were any cancellations at all, given that is was presumed that the timetable was being introduced as a result of a level of sickness above normal level amongst conductor guards. The second shock was just how drastic these cuts were on some routes, whilst others were almost completely unaffected.

We are used to hyperbole from MPs but Helen Hayes, MP for Dulwich and West Norwood, is probably not exaggerating at all on her website where she publishes the contents of a letter to Rail Minister Claire Perry. In it she states:

The services to be cancelled are heavily concentrated in South London, including the vast majority of the peak hour commuter services running from stations in and very close to my constituency – North Dulwich, East Dulwich, Peckham Rye, Denmark Hill, Gipsy Hill and Crystal Palace as well as the withdrawal of all services to London Bridge from West Norwood and Tulse Hill.

This is effectively the collapse of commuter rail services in my constituency, and coming on top of more than 18 months of appalling service levels, cancelled trains, short trains and overcrowding as a consequence of the earlier phases of the London Bridge works, it is completely unacceptable.

Actually, one could argue that Helen Hayes has in some ways understated the cuts, as at stations outside her constituency the situation can be even worse. Birkbeck, for example, admittedly already the fifth least used station in London, now has no trains at all on Mondays to Fridays. As it is also devoid of a train service on Sundays, it now only has a service on Saturdays.

To try and give some context as to the scale of the timetable cuts, below is detailed the entire peak service to and from London Bridge, from three inner London stations:

To London Bridge:
from North Dulwich: 6:53, 7:27, 8:27, 9:28
from East Dulwich: 6:55, 7:30, 8:30, 9:30
from Peckham Rye: 6:58, 7:33, 8:33, 9:34

From London Bridge: 16:10, 16:40, 17:11, 18:08, 18:51

and that being pretty generous with a definition of what constitutes a peak period.

Yet, incredibly, the Tattenham Corner branch, not known for the hordes of people that use it, emerges unscathed. Furthermore the Tattenham Corner branch will, of course, continue to have a full Sunday service when, at times, the number of coaches on the train will exceed the number of passengers.

East Dulwich has over 2 million passenger journeys a year. You can’t easily absorb that number into a local transport network without further consequences. By means of a contrast, the seven stations on the Tattenham Corner branch combined have about the same number of passenger journeys each year.


The cuts might initially seem unbelievable. It is possible to see though what Southern were thinking – even if one doesn’t agree with it. There is a perception that in inner London alternatives are available, which means either other stations nearby, other services from the same station (as at Peckham Rye or the West London line which is also having its Southern service withdrawn) or there is a decent bus service in the area. The problem is that these alternatives are generally full up. Even where the passenger involved has an alternative (such as London Overground) it could well mean that another person further down the line will not be able to physically board a train.

One can also see that cutting services even further into London Bridge would help maintain some sort of decent service into London Bridge for services via East Croydon without the above (and other some other) stopping services getting in the way and causing congestion in the London Bridge area. It is also the case that by cutting out services in inner London that cross a lot of flat junctions one is providing more opportunity for other trains to run to time.

Nevertheless, the emergency timetable is expected to cause near paralysis in some areas with the only saving grace that the school holidays are almost upon us, which will not only mean schoolchildren not wanting to travel by train but also more parents taking leave from work to go on the family holiday. And if the emergency timetable is still in place in September what can we expect then?

Rolling Stock

Anyone who spends their days problem solving in the real world will be well aware that it is often the case that one prominent problem often hides further problems lying below the surface. And so it is that we would argue that rolling stock – or the lack of it – is the hidden problem on Southern. With the current problems in such prominence, it is easy to forget that earlier this year trains were being cancelled on Southern due to a shortage of rolling stock. It is also notable that even with the existing level of cancellations across Southern we sometimes see trains short formed. Note that Helen Hayes (above) refers to short trains.

As with all TOCs, stock utilisation on Southern Railway is pretty high. The rolling stock plan, almost entirely controlled by the DfT regardless of what they might claim, has many problems. Some of this is a knock-on effect as a result of delays to Network Rail electrification. It was expected, even on the revised schedule after the two year delay in ordering Thameslink trains (down to the DfT), that many more of the class 700 Thameslink trains would be in service and fewer, if any, of the old class 319 Thameslink trains would be running. More relevantly to Southern, they would have expected by now to be much less reliant on their metro class 455 stock and have more reliable (and newer) stock such as the Bombardier class 387 Electrostar to replace it. All aggravated, if reports are correct, by the fact that the 377s inherited from First Capital Connect on Thameslink have required more work to bring them up to scratch in reliability terms than expected. This has had a knock-on effect on maintenance and improvement work on Southern.

What appears to be surprisingly lacking in the emergency timetable is a commitment to running longer trains for the busy trains that are running and can be lengthened. This is something you would expect if the problem was only a shortage of train crew. There is nothing definite to substantiate this assertion about a lack of rolling stock but there are strong suspicions.

It might be informative to note that cutting back on the lightly used Tattenham Corner branch during the peak period does not save on any rolling stock. The train divides at Purley and in the event of the service not running to one branch all of the train is sent down the branch still running. The situation is also complicated by some rolling stock stabling at Tattenham Corner.

London Bridge

Whilst the problems originally encountered at London Bridge are generally resolved, it is still the case that there is only one through line in each direction just north of New Cross Gate. This severely affects flexibility when things go wrong. This particular situation will not get better until January 2017 when an extra down track going through the Bermondsey Diveunder will be brought into use and not fully resolved until 27th December 2017.

Ongoing signalling issues

An issue that seems to be largely forgotten or underplayed is signalling, which is still causing problems. Some of this is inevitable, given the old signalling still around on the Brighton Main Line, but there are still some delays due to “a signalling issue in the London Bridge area” which makes one suspect it is the new equipment that is still playing up.

Signalling is, of course, entirely down to Network Rail but in the minds of most of the public it is inevitably Southern who get the blame. This is a case of “don’t shoot the messenger” but, as always, the quality of information provided by the TOC becomes an issue.

Reactionary Delays

Another problem that Southern Railway is particularly prone to, more than any other train operator, is reactionary delays. Put simply when something goes wrong on Southern it almost always has knock on effects that are at least double the original problem. This is largely down to the desire to squeeze out the maximum capacity theoretically possible out of the network leaving very little margin for error or correction. Those flat junctions and routes crossing each other clearly don’t help.

The effect of reactionary delays on Southern is to turn minor problems that would quickly settle down of their own accord into major ones that have to be managed. This is a challenge at the best of times but clearly more so when you get too many reactionary delays occurring at the same time.

Dwell time

Dwell time at stations is a major modern problem for all sorts of reasons that would take a separate article to describe. For our purposes it is sufficient to point out that when trains are very crowded it goes up dramatically as people struggle to get on and off trains. The cumulative effect at each station can cause significant delays which of course have their own knock on effects.

A particular problem with dwell times occurs at East Croydon. It is particularly serious for up slow trains which in practice all use platform 4. When trains are overcrowded a train can be scheduled to have a one minute dwell time which can easily turn into three minutes in reality. Putting it another way, when there are dwell time issues, a train can arrive on time but depart two minutes late.

Conductors’ dispute

Finally as readers will probably be aware, there is a conductors’ (guards) dispute on Southern Railway. There is also a shortage of conductors but it is fairly clear from the Parliamentary Select Committee meeting to oversee Southern Railway that Southern don’t want to fill these roles but instead rely on a slight increase of Driver Only Operation. Whether this is planned and included in rostering or unplanned and only intended to cover situations when no guard, for whatever reason, is not available, is not made clear.

The dispute does not explain the delays and cancellations on a daily basis. This, Southern argues, is down to “an unusually high level of conductor sickness”. Southern Railway are clearly suggesting not all of this is entirely genuine but many, including, apparently, the Select Committee do not appear to be entirely convinced. Southern’s argument appears to rely on believing there is an otherwise unexplained increase whereas, for others, stress caused by the ongoing industrial action and generally less friendly working environment seems to be an entirely plausible explanation.

What of the immediate future?

If our analysis is correct, possibly the worst nightmare for Southern Railway is that sickness amongst conductors dramatically falls to below normal levels and there are still a lot of cancellations on their services. Either that or they impose new conditions on August 21st, as planned, but still have a lot of cancellations. This would leave them with a lot of explaining to do. Of course, if sickness levels fell and the service miraculously ran more-or-less to time with no or very few cancellations then Southern would, in the eyes of some, be vindicated.

Meanwhile, for six weeks at least, it looks like the emergency timetable will come into force and no doubt there will be many questions in parliament.

What of the long term future?

It is clear that many are clamouring for GTR to be stripped of their franchise such as Chuka Umunna (Labour) and Chris Philp (Conservative). It is also clear that GTR has defaulted on their franchise requirements in multiple ways, but the DfT have no intention of acting to have Southern (or all of GTR) stripped of it.

Apart from the challenge of replacing Govia there is a question over whether anyone would want to take this franchise on – and if they did they would probably want to be well rewarded. It would only be a distraction and not get down to the heart of the problem. Besides, incredible as it may seem to some, there is a feeling that no-one else would be able to do any better. Indeed, if one accepts that the conductors’ dispute is only one part of the problem, it is hard to see that much can be done. If the high level of sickness is genuine then there are no instant fixes. It is also hard to see how settling with the RMT would actually help the situation given that they, at the time of writing, have not called any more strikes.

Nevertheless, if things turn out badly when Thameslink fully opens in December 2018 it is hard to see how a Secretary of State could avoid terminating this franchise early – or at least part of it. There now appears to be cross-party consensus that the best thing to ultimately happen is for TfL to take over in the Southern Railway metro area. If improvements then didn’t materialise, at least the Mayor would be accountable for what is going on.

Meanwhile we are quite sure this will not be the last we write about this. As ‘Meltdown Monday’ approaches we are left wondering what the next instalment will bring.

Thanks to the considerable assistance provided by ngh and Graham Feakins in the preparation of this article.

Like what you read? Buy our magazine

Find out more about the challenges of building Crossrail, TfL’s quest to take over London’s suburban railways and Sir Peter Hendy’s mission to change the perception of Network Rail in Feeding the Monster, our latest issue. Buy it now

Written by Pedantic of Purley