Railpolitik 2017: how to improve our railway operation for its customers

What follows is a summary of a speech I gave at the Modern Railways Golden Whistles Awards in January 2017. This covered five interlinked subject areas.

Nationalisation

How much of our current railway is actually privatised? While the question of nationalisation frequently stokes political debate, I think this usually gives off more heat than light.

Network Rail is effectively nationalised, and is closely controlled by the Government. Operating concessions, such as those let by Transport for London, are tightly specified by a public sector client and let to a private sector delivery organisation. Most franchises are also tightly specified by the Department for Transport, with limited risk transfer to the private sector. The one exception is rolling stock, whose funding and ownership remains predominantly in the private sector.

I believe there is now far greater Government control over decision-making than there ever was in the days of British Rail. This is an issue on which we really need to educate the mainstream media – I believe the railway can flourish in either the public or private sector, and we need to move the focus away from this issue to those which really matter.

 Devolution

Transport Secretary Chris Grayling’s decision not to devolve control of South East London Metro services to Transport for London in the next franchise bucks a recent trend towards devolution.

There are many successful examples of where it works well – the Transport for London operating concessions (London Overground and TfL Rail) have been outstanding successes, and Merseytravel has great strategy and vision for rail services in Liverpool, with excellent delivery by operator Merseyrail. In spite of the current issues it is facing, devolution of the ScotRail franchise has also been largely beneficial, with strong client leadership from Transport Scotland. Local authorities in the West Midlands have developed a good vision for local rail services around Birmingham, although it seems the prospect of full devolution of control will not now happen.

In contrast, the Govia Thameslink Railway operation has all the symptoms of a badly specified franchise. It is simply too big, but it should be possible to break it down into smaller businesses covering the metro routes and Thameslink/main lines; this would also improve train crew arrangements.

I believe the London Overground-style devolution approach could be made to work for the South Eastern, Southern and South Western metro routes. A frequent counter-argument is that of ‘democratic deficit’ regarding the management of services outside the Greater London boundary, but this could be handled through governance arrangements on the client side. The Overground approach could bring rationalisation of routes to improve frequencies, with easier interchanges and clearer route branding.

Where devolution is concerned, it seems to me that short-term political considerations have trumped sound policy and the needs of Londoners.

Vertical integration

At the same time, Chris Grayling announced a plan for ‘virtual vertical integration’, bringing operations and infrastructure closer together. The railway has some experience of this with the Wessex Alliance and the current arrangements in Scotland, but as yet there is no detail as to how it might work on the first two franchises it will be applied to – South Eastern and East Midlands. Timescales to develop this work ahead of the issue of the franchise invitations to tender are tight.

So how might it work? There are inherent differences between the priorities of infrastructure owners and train operating companies (TOCs). Infrastructure assets are long-lived, whereas TOCs need to focus on revenue generation over the shorter-term length of their franchise. Infrastructure is capital-heavy, whereas TOC owners tend to be capital-light. Infrastructure risks are partially unknown, and can prove to be very expensive. I believe that the industry still does not have a proper handle on asset condition, and therefore this risk remains difficult to assess. TOCs are customer-facing service businesses, but infrastructure companies concentrate on assets and tend to be funder-facing. Given these differences, aligned incentives and culture are the only way to make these arrangements work.

The deep alliance arrangement between South West Trains and Network Rail was based only on operations and maintenance costs, whereas the ScotRail Alliance includes renewals (albeit tightly capped). I believe the Wessex alliance led to a greater focus on infrastructure problems, meaning less energy was put towards customer-facing initiatives and performance.

We have also seen that TOCs and Network Rail have different organisational cultures and governance styles, so the decision-making processes and management of relationships can be challenging. A further major challenge, especially for the East Midlands arrangement, will be the operating area, particularly where other operators are involved. Generally, infrastructure requires a long concession term, say 30 years, to correctly steward and develop assets, while operators need a much shorter term (up to 10 years), otherwise revenue risks become unmanageable.

The success of an alliance ultimately comes down to the creation of an effective pain/gain share mechanism to incentivise the parties’ behaviours. Creating such a model is critical to the policy’s success but will be extremely challenging.

 Franchising

Is franchising in crisis? Some commentators are predicting its imminent demise, and I would agree that the market is weaker today than at any time since privatisation. Margins for operators have become unsustainably low, and the risk/reward ratio and reputational risk are not conducive to bidders. The owning groups have insufficient balance sheet strength for the model which is emerging, while invitations to tender have grown more and more complex, and will only continue to do so given the emphasis on virtual vertical integration. Few credible new players are emerging capable of bidding on their own for large franchises – National Express’s sale of c2c to Trenitalia doesn’t exactly strengthen the market, and there is an increasing trend for companies to form joint ventures. Given all that, something must change if franchising is to survive.

I would abandon the move towards vertical integration, accepting that infrastructure and train service provision are quite different. I believe we need smaller franchises, big enough to have economies of scale but small enough to be manageable, and these should align with one clear market; franchise terms should not be too long in order to keep revenue risk manageable. We should accept that tight specification and strong client teams are needed to secure an acceptable quality of service in the interests of passengers, and this should be driven through invitations to tender and incentives which focus on improving quality and consistency. On the infrastructure side, I support Mark Carne’s moves for route-based accountability within Network Rail, and believe there should be full devolution of performance regime impact to Routes, with freedom for Route managing directors to manage and real accountability for the results.

 Performance

Operational performance peaked in autumn 2009, before plateauing for the following three years, but since then it has been in decline. A busier network and greater delay following each incident is often claimed as the cause, but network busyness has actually been relatively static in the past two years, yet the decline in performance has continued. Looking at the past year, the London and South East operators and those on the East Coast main line have struggled, but the West Coast main line has bucked the trend and TfL Rail is a clear outperformer, showing what can be done.

The current difficulties on GTR are obvious, but there are other issues. Is performance getting the attention it deserves? Is there a tension between infrastructure cost reduction and performance improvement? Some areas are working better – concessions such as TfL Rail, those smaller operators with a tight geographic focus and a close relationship with their lead NR Route, and areas which have successfully forged a common purpose, such as the West Coast main line.

How do we tackle this? I would split GTR up, and would generally move to split big TOCs into smaller units with a stronger incentivisation and focus wherever possible. For areas of sustained underperformance, we need to create a burning platform to force efforts to improve matters, such as on the East Coast main line. Areas of sustained poor infrastructure performance should be identified, with funding put in place to produce a long-term fix.

Looking ahead to Control Period 6, we need to declutter the agenda and get industry and politicians to focus on two key areas – growing capacity and improving performance. I would add sorting out fares and ticketing, as this is another key issue, and is not that difficult to fix if there is suitable determination.

As a final note, I would urge the industry to abandon the public performance measure (PPM). It is simply not fit for purpose, and has encouraged a move away from right time running. As an industry, we need to change our focus and get to grips with small repeated delays as a start to improving performance. Punctuality should be measured internally on right time performance at key junctions, and externally on right time performance at calling points, weighted by volume of passengers.