By Peter Mackie
Professor of Transport Studies, Institute for Transport Studies (ITS), University of Leeds
It is a feature of many big transport projects that they go through massive wobbles. The Third London Airport, the Channel Tunnel, HS1, Crossrail are all examples and the very name Thameslink 2000 is a reminder of the syndrome which HS2 is now experiencing.
My overall position having had the benefit of membership of HS2’s Analytical Challenge Panel until March 2012 is that HS2 is not a stupid ill-conceived vanity project but neither is it a game changing transformational project for the economy. It is best viewed as a major transport project with wider economic, social and planning impacts. It is worth doing provided the price is right, and that is not yet clear. It absolutely must be considered on its merits and the starting point for doing that is the costs and benefits to the transport system, and then to the spatial economy. It is actually pretty implausible that HS2 on its own could fundamentally change the North/South divide. That would require some bigger decisions such as relocating Westminster to Birmingham! But that is not to deny that HS2 could make a very useful contribution to improving the connectivity of the core cities which are the best prospects of engines of growth we have got.
Britain has a lot of infrastructure challenges not just in the transport sector but in energy, water, flood protection and urban renewal. Those who say we would be better off with an Infrastructure Planning Commission to help mediate these choices are undoubtedly right, but in its absence the best we can do is to imagine that the x billion committed to HS2 would otherwise be committed to useful public infrastructure, some but not all in the transport sector, with a decent social rate of return. So therefore the fundamental requirement ahead of the Hybrid Bill is to assess the case for HS2 starting from the bricks and mortar of the transport business case. One of the strengths of the Public Inquiry system is that the scheme promoter is required to state its case and be subject to cross-examination. With that in mind, here are six questions on which clarity is going to be required.
Question 1: What has happened to the costs and what can reasonably be expected to happen?
My appreciation is that the costs were based on the unit costs for HS1, the most expensive railway in the world, and then loaded with a further 66% mark up for optimism bias. That produced the 32bn number. Then on top of that there were some subsequent specification changes, for example the link to Crewe, which pushed it up further, maybe to 35bn. Then it was decided that we wanted the 95% probability value of the costs rather than the central 50th percentile pushing the headline number up to 42 bn. But there are questions about this – surely if you move to the 95th percentile you should reduce the optimism bias mark up because you are taking a far more conservative position. In any case for the cost-benefit analysis central case, what is required are the central values – the expected 50th percentile costs, revenues and benefits – with a risk analysis around the central case. This needs explaining very clearly: the system seems set up to produce a very high cost forecast which the contracting system will then magically just manage to achieve.
Question 2: How robust are the traffic splits between classic and high speed rail at equal fares?
There seems to be a bit of a conundrum about HS2. In some ways it serves some places off the line such as Newcastle, Liverpool and Preston better than several of the places on it. We seem to be unable to connect HS2 properly to the main stations at Birmingham, E Midlands, Sheffield or Leeds. Realistically this is going to be a middle class project so where potential users live is important. How will the respectable burghers of Ilkley, Bingley, Dore, Fulwood, Bingham, Edgbaston, Wilmslow, Prestbury and so forth choose to get to London and how is a bracing walk from say New Street to Curzon Street on a February day with a suitcase reflected in the choice process?
Question 3: How does HS2 scrub up in financial appraisal terms?
There are numerous aspects to this. One is what the trade off is between revenue and benefits in the fares policy. There seems to be a curious dichotomy here. On the one hand the value of a step change in journey times is stressed. On the other, when it comes to converting that into revenue through a premium fare, the analysts seem to run into modelling problems. I can accept that representing yield management systems in models is difficult, but if we are unable to say how much of the user benefit from reduced journey times we could convert into additional revenue, we are in trouble. More generally, we just have to know the expected impact on the finances of the railway industry as a whole. If the Government cannot build HS2, endow it to the industry and operate the industry within the same financial envelope as without HS2, that is an acid test failed in my book. The crucial financial test is not what the HS2 franchise can be sold for as a job lot – Lord Heseltine has suggested £10bn – it is the impact on the rail financial envelope as a whole. Obviously answering that question implies the need for the Government of today to take a view on the future organisation of the railways in terms of infrastructure, franchise and open access.
Question 4: What is the social and economic value of increased capacity delivered by HS2?
I think a big criticism of Government as opposed to the DfT, still less HS2 Ltd, is the siloed approach to the strategic case for this scheme. It is very difficult to evaluate the benefits of more capacity on the WCML without a plan describing how that capacity will be used. My appreciation is that we have a national problem of accommodating serious population growth in South-East England without much idea how the infrastructure is going to deliver it. HS2 has the potential to help solve some serious social problems but no-one seems to dare to think of HS2 as more than a transport project. Surely a more prescriptive less dirigiste approach to planning is required in which we say for example that we are going to exploit HS2 by building Milton Keynes 2, and developing the areas around Old Oak and Park Royal in London and Birmingham International in certain ways. It is really difficult to make the strategic case for HS2 without a spatial strategy.
Question 5: What exactly is not yet in the HS2 Benefit : Cost Ratio and when and how will it be decided if the Value for Money category of HS2 is affected by the omitted items?
The promoters have made much of the delays to users of existing lines under maintenance if HS2 is not built. It is not yet clear how many of those delays will happen anyway but it is also unclear whether the HS2 BCR yet includes delays during construction of the scheme itself, particularly but not exclusively at Euston. More generally we hear that the Environmental Statement will run to 55,000 pages but more interesting is how and when that will be summarised and brought into the overall assessment of social value for money according to Green Book guidance.
Question 6: What weight can be attached to GVA methods in general as indicators of strategic impact?
The details of the methodology used in the KPMG report have been the subject of evidence to the Treasury Select Committee by KPMG and Professors Overman and Graham. Without entering the detailed fray, there are some general points about this which have an importance far beyond the HS2 project.
The cost-benefit analysis framework, represented in the Green Book and WebTAG, has a history of forty years behind it. It is not to everyone’s taste (see for example letter of Alan Wenban-Smith to Local Transport Today of 15/11/13). But it has some strengths :
- It provides a coherent framework within which the assessment of transport projects takes place
- It provides a basis for taking the results of models required for traffic forecasting and using those results in an internally consistent manner for project appraisal
- It operates within a defined set of assumptions relating to the macroeconomy which are debatable but are at least a level playing field
- It relies on values for travel time, safety, overcrowding etc which are evidence based even if, just like all the other parameters in the process, the quality of the evidence is open to challenge
- It starts from the proximate impacts of projects on travellers and transport operators and then works out from that to the spatial and additional wider economy impacts
- It should be comprehensive in nature and provide the basis for bringing together the economic, social and environmental impacts in one place.
My appreciation of Gross Value Added metrics in general – as opposed to particular applications — is:
- They are relatively new and there is limited experience of using them
- They focus on the GDP effects of an intervention but actually in a sector like transport this is inherently difficult to forecast. For example, how do business travel time savings, commuter time savings, overcrowding, reliability and safety benefits convert into GDP effects? This is not an easy question because these things overlap the ‘real’ and ‘wider’ economy but unless it can be answered, the changes in accessibility at zonal level cannot be relied upon
- The implicit macroeconomic model on which the results are based is not highly transparent. For example, where improved accessibility is expected to increase output at regional, local or even national level, it is not clear how the net value of the increased output after taking account of input costs and displacement from other sectors is computed
- They clearly omit impacts which are non-market but included in the CBA
Overall, if GVA methods are to find a place as a reliable component in transport assessment, work is needed to build the appraisal rules within which they can be used. In the particular context of HS2, the obvious question is ‘ How could the GVA impact of HS2 be of the order of ten times the benefit impact in the CBA?’ Some degree of sense-checking and rationale needs to be brought to this. For the moment, the combination of the direct user and operator benefits, the induced land use change impacts and the additional wider economy impacts is likely to provide a more robust appraisal metric.