Road appraisal makes carbon dioxide uniquely insignificant. Why? And what to do about it?

The decisive current calculation for carbon assessments of road schemes is a unique ratio:

The estimated additional carbon resulting from the scheme

divided by

The total carbon emissions in the economy

The implications of this emerged last month in Lynn Sloman and Lisa Hopkinson’s thoughtful and well-sourced report, which concluded:  “[The Roads Investment Strategy] RIS2 will make carbon emissions from the Strategic Road Network (SRN) go up, by about 20MtCO2, during a period when we need to make them go down, by about 167MtCO2. This increase in CO2 from RIS2 will negate 80 per cent of potential carbon savings from electric vehicles on the SRN between now and 2032.” [1]

Roger Harrabin, the BBC’s environment analyst, quoted a brief unexplained rebuttal from the DfT saying “This assessment is wholly incorrect.” [2] Part of the disagreement will be about which carbon impacts to include, and what estimate to make for induced traffic, ie the numerator in the ratio above, which Highways England, in my judgement, has systematically underestimated. I’ll write more on this in a future column. But the denominator, usually ignored, has a much greater effect. 

The basis for the ratio seems to be an interpretation of an official statement set out in 2014: “The Government has an overarching national carbon reduction strategy… The Government is legally required to meet this plan. Therefore, any increase in carbon emissions is not a reason to refuse development consent, unless the increase in carbon emissions resulting from the proposed scheme are so significant that it would have a material impact on the ability of Government to meet its carbon reduction targets.” [3]

At first casual reading this statement seems unobjectionable, but Sloman and Hopkinson show – with eye-opening quotes from a dozen or so Highways England schemes, on page 12, table 2, of their report – that this principle is used in practice as the basis for a statistic in a more or less standard form of words: “emissions arising as a result of the scheme represent less than 0.1 per cent of total emissions in any five-year carbon budget in which they arise… [therefore] the greenhouse gas impact of the scheme would not have a material impact on carbon reduction targets as set by the UK Government.” 

If the 2014 principle is interpreted in this way, it must always, inevitably, become a formula for dismissing the importance of carbon, rather than enhancing it.  

The denominator is the total volume of carbon dioxide emitted in the whole economy, This is inherently a number several orders of magnitude bigger than the conceivable output from any one particular scheme. So any revision to modelling, induced traffic, classification, or what to include, can only make an insignificant difference to the conclusion. The small numbers result predominantly from the choice of denominator, not from the scheme-specific modelling of the numerator.

The base case [for road scheme appraisal] would have to be traffic trajectories compatible with meeting the carbon objectives. This would have drastic effects on the business case for schemes.

As far as I know, the use of a denominator of the ‘whole economy’ type is unique to carbon in road appraisal. 

It is not applied to any other effect of a road scheme. Consider if one calculated – say – the total number of jobs generated by a specific scheme, as a proportion of the total jobs in the economy; or the number of minutes travel time saved by a particular scheme, as a proportion of all the time spent on everything in the whole country; or the change in the total number of lives lost in crashes associated by a particular scheme, as a proportion of all deaths from all causes in the country. 

Any such calculation will always come up with a number in the order of an impact of less than 0.1 per cent, and be open to words like ‘immaterial’ or ‘insignificant’. But it is immediately obvious that this does not in the slightest prove that the effect is unimportant by comparison with other methods of increasing employment, or saving time, or saving lives. 

The denominator is peculiar, I think, to appraisal of road construction. There are many transport carbon-related decisions taken by, or of great interest to, Government which are not dismissed by use of this denominator. The choice by individuals whether to use electric or internal combustion vehicles is the subject of great policy interest, even though it is clear that such a choice, for each individual, or each transport operator, or each fleet purchasing manager, or each local authority, would equally be described as a decision having “no material impact on the ability of Government to meet its carbon reduction targets”. 

It is not how decisions are taken on building a new power generation plant, or a scheme for insulating houses, or a scheme for planting trees. In all of these cases the principle is that relatively small separate effects can add up to important total effects, relative to the sector or type of scheme. 

So what special reasons might there be to choose a unique type of comparator for carbon as applied to road appraisal? It is true that there is one widespread and powerful argument for treating carbon uniquely, namely that it is so important, the survival of human societies being at stake, that it should be taken outside the normal assessment of trade-offs in a benefit-cost calculation, and treated as a prior condition or filter. However, it would be completely perverse to use an argument of the unique importance of carbon as a justification for use of a ratio that must make it appear uniquely unimportant. 

The argued justification for using this indicator seems to be that the Government, because it is legally required to meet carbon targets, can be relied on to meet its promises. It has sufficient tools at its disposal (pricing, regulation, market intervention, investment etc) that it can do so. Therefore the trajectory to meeting the commitments can be taken as a matter of core policy that  will happen. To the extent that there are any negative consequences of road building, they can and will, by definition, be offset by other policies. The target will be met because it must be met, somehow, and therefore the actual method or cost of offsetting these effects does not need to be included in the scheme appraisal. 

I understand that it would be possible in principle to offset any additional carbon consequences of a road scheme: for example, there could be a carbon-based road user pricing system charged at a level that would reduce traffic, rather than inducing more. But that is not policy. A core feature of all appraisal carried out by the DfT’s TAG procedures, is that the ‘base’ case against which variations are tested always explicitly excludes any possible future policies or projects that have not been defined in detail, committed in law and funding. 

Only specific already-guaranteed future changes can be included in the modelled forecasts, with their associated effects on transport costs, choices, traffic volumes, etc. Uncommitted future policies only appear, if at all, in sensitivity tests, policy tests, or subordinate scenarios.

Therefore I think the implication of adopting the argument ‘the target has been promised therefore the trajectory will happen’ is that it must logically also be applied to the base case for all scheme appraisal. The base case would have to be traffic trajectories compatible with meeting the carbon objectives, and therefore would need to include such changes to driving costs, regulation, restrictions or other factors that could bring these about. This would have drastic effects on the business case for schemes.

The fact is – I don’t think Government would disagree with this – that the current mix of policies, prices and projects is insufficient to meet the current carbon targets, more so in the transport sector than any other, and there will need to be initiatives that have not yet been determined. 

To use this ratio in advance of defining and activating those initiatives seems to me to run counter to the logic on which all official recommendations on road appraisal are based. There is manifestly still an important unresolved argument to be had about the true carbon effects of the road programme, but a necessary condition for that argument to be meaningful, I would say, is formally to retire this unique denominator, which so restricts the scope of decarbonisation discussions. 

Phil Goodwin is emeritus professor of transport policy at both the Centre for Transport and Society, University of the West of England, Bristol, and University College London. Email: philinelh@yahoo.com

Notes: 

[1] https://www.transportforqualityoflife.com/u/files/The%20carbon%20impact%20of%20the%20national%20roads%20programme%20FINAL.pdf

[2] https://www.bbc.co.uk/news/science-environment-53353258

[3] https://www.gov.uk/government/publications/national-policy-statement-for-national-networks para 5.18