Written evidence from the Independent
Transport Commission (TE 27)
This paper reflects recent research by the Independent
Transport Commission and is relevant to the following lines of
enquiry outlined in the call for submissions:
(i) How should the balance between revenue and
capital expenditure be altered?
(ii) Are the current methods for assessing proposed
transport schemes satisfactory?
THE CASE
FOR TRANSPORT
INVESTMENT: A REVIEW
OF SOCIAL
COST BENEFIT
ANALYSIS
1. SUMMARY
1.1 Overview:
The Independent Transport Commission (ITC)
believes that transport is good at showing returns
on investment. The combination of stringent tests prior to spending,
and before-and-after measurements of delivery means that the Treasury
can be confident every penny spent on transport earns good returns.
This strong sentiment led to this submission on Social Cost-Benefit
Analysis (SCBA), the current method of spending appraisal.
1.2 The Strength of Transport's funding case:
Social Cost-Benefit Analysis came into transport
in the 1960s, to appraise spending on road building. Today it
is used for all transport investments in excess of £5 million
and many smaller scale local investments. The basic principles
underpinning SCBA have not changed: it measures what we deliver
to society in return for expenditure on transport. Unlike other
spending areas, transport currently only claims as "benefits"
those things directly part of transport - such as increased revenues,
reduced costs, reduced accidents, and travel time savings. All
those extra impacts which transport schemes have on jobs, education,
health, society count as "below the line" in transport,
but they are often included "above the line" in other
spending areas. This rigour often means that Transport investment
plans understate benefits that other Departments would overstate.
In the October review, therefore, it is critical that Treasury
demands the same rigorous analysis in other spending areas.
1.3 Opportunities for Improvement:
The UK has a well-developed methodology for SCBA
and strong expertise in its application. That said, the Independent
Transport Commission proposes improvements to the SCBA method:
better consideration of non-monetary impacts; greater room for
citizen-led prioritisation of policies; more explicit risk assessment
around forecasts given their long timescale, and an audit trail
between the value-for-money assessment and political decision-making.
1.4 Recommendations:
1.4.1 What should be done? The ITC recommends:
In the short term, the following aspects of the
technique should be reviewed:
1.4.1.1 Carbon savings should, given the difficulties
of monetising their effects, be assessed on a least-cost rather
than a benefit-cost basis.
1.4.1.2 As not all time savings have the same
value and as they are sometimes used as proxies for local spending
and increased employment, they need reconsideration.
1.4.1.3 The costs and benefits of transport investment
differ depending on social group and geographical area. Recent
developments in mapping social and local transport impacts should
be used.
1.4.1.4 Given the prospect of cuts, cost-benefit
analysis should be adapted to deal with infrastructure contraction
as well as expansion. For example, the narrow terms used by Dr
Beeching to decide railway cuts in the 1960s failed to recognise
the varied impacts of closure.
1.4.2 In the longer term, transport
appraisal needs to be subject to a root and branch review that
would incorporate approaches used in other industries and countries.
The current moratorium on major schemes presents a unique opportunity
to undertake a fundamental review without prejudicing investment.
2. BACKGROUND
2.1 The ITC, the UK's only independent transport
commission, held a discussion evening on 17 June. Chaired by
Willy Rickett, previously the author of Government's "10
Year Plan" transport development strategy, and hearing from
speakers Stephen Glaister, Elizabeth Gilliard and Andrew McNaughton,
the group debated the subject "How would you spend £50 billion
on improving Britain's Infrastructure?". A summary of that
discussion is available on the ITC's website entitled "Rethinking
how we spend money on Transport".
2.2 Some significant part of the discussion concentrated
on the method of how we appraise and justify major transport spending
decisions. Some people argued strongly in favour of the rigour
of the existing Social Cost-Benefit Analysis (SCBA) method; while
broadly supporting the approach, others seemed to be more sympathetic
to adopting a modified version of the system which takes more
fully into account the value of reducing CO2 emissions
and some modifications to the current treatment of time savings,
while some felt that the method was fundamentally flawed. But
equally arguments were made that we need a methodology that could
be used now. At the end of the discussion, the Chair attempted
to assess the mood of the meeting and was surprised that only
a modest array of hands were raised in unequivocal support of
the current SCBA methodology.
2.3 Only two days' before the ITC meeting, the
new Secretary of State speaking at a London First event "The
Transport Priorities of the New Government" had displayed
similar reservations about the rigours of the SCBA method, indicating,
in an answer to a question, that he would also wish to see local
issues appropriately taken into account and a requirement to distribute
investment around the country as additional criteria for informing
transport decisions. So SCBA has very recently been exposed to
greater public scrutiny.
3. THE NEED
FOR APPRAISAL
3.1 Where consideration is being given to investing
in major transport infrastructure projects, there is a need for
a method that will help both to determine whether a scheme offers
"value for money" and is cost effective, and to assist
in prioritising among a long list of potential schemes - some
very different in nature (e.g. motorway building vs. rail
electrification) - given that funds are invariably limited.
3.2 In the commercial sector these requirements
can be achieved through established methods of financial appraisal,
but in the case of transport where much investment involves the
public sector, the SCBA method has evolved in recognition of the
fact that:
3.2.1 The costs and benefits of both transport
investments and operations are not confined to those who provide
and use them, but affect non-users and the community as a whole,
to a greater degree than almost any other sector of the economy.
The clearest examples are noise, congestion and atmospheric pollution.
3.2.2 The use of the majority of transport infrastructure
- the road network - is not allocated by the market in this country,
but is provided free at the point of use. This means that the
user benefits cannot be assessed directly through willingness
to pay for the service (other than crudely through the fuel duty
and Vehicle Excise Duty).
3.2.3 Even where services are provided by the
market and directly priced (e.g. public transport fares), relatively
straightforward pricing structures are often used and so provide
an imperfect measure of the benefits that users derive from the
service.
3.2.4 Major transport infrastructure is costly
and has a long operational life - hence the need for a comprehensive
and rigorous appraisal of the investment.
3.3 So in light of these considerations, some
kind of evaluation method is required.
4. THE EMERGENCE
OF SCBA
4.1 Although the principles of SCBA can be traced
back to 19th century welfare economics, the method was first developed
and used by the US Army Corps of Engineers to prioritise investment
in the irrigation schemes that they were building as part of the
New Deal in response to the Great Depression.
4.2 It was taken up in this country in the 1960s,
where it was applied to transport schemes through a retrospective
analysis of the M1 motorway (Beesley) and the Victoria Line (Foster
and Beesley). It was then adopted by Ministry of Transport in
the 1960s and applied to a number of small projects in the roads
programme. Since then it has been applied both to major road
and rail schemes, and is formally embodied in the methods required
for appraising transport investments in excess of £5 million
(i.e. as set out in NATA in England and Wales, and STAG in Scotland).
5. SCBA TODAY
5.1 The basic principles underpinning SCBA have
not changed. The core elements of the process are: 1) Specify
options; 2) Identify the costs and impacts; 3) Predict the costs
and impacts over the life of the project; 4) Monetise those impacts
for which sufficient data and justification exists; 5) Discount
the costs and impacts to obtain present values; 6) Conduct sensitivity
analysis and 7) Compare the net present value of the alternatives.
5.2 Whilst in its early applications SCBA focused
almost solely on time savings, vehicle operating costs, fares
and accident reduction as the monetised benefits of a scheme,
a range of additional measures are now monetised including improvements
in journey experiences at interchange and noise. Values for impacts
such as agglomeration benefits and other wider economic impacts
are being developed as are estimates of air pollution damage costs
and climate change emissions. There remain a number of factors
that sit outside of the SCBA calculation, but which form part
of the overall assessment of the value for money of a project,
including impacts on the built and natural environment, water
quality and regeneration benefits. These other factors are all
presented as part of a wider Appraisal Summary Table under the
New Approach to Appraisal (NATA).
5.3 In studying the application of this framework
to the 1998 road review of 68 trunk road schemes, a study found
that the Appraisal Summary Table information appears to have corrected
for the traditional in-built bias towards monetised impacts and
that the scores for noise, landscape, heritage, reliability and
regeneration were "all significant explanators of the decision
pattern".[59]
5.4 The project appraisal period for most projects
has, in the recent past, been extended to 60 years since the discount
rate has been reduced to between 3 and 3.5% from 6%. In addition
to this, evidence of a systematic bias in cost underestimation
for projects has led to the introduction of an "optimism
bias" uplift to projected costs which varies with the nature
of the project and the degree of risk mitigation identified, but
can be as high as 66% at pre feasibility stage.
6. PROBLEMS WITH
THE CURRENT
METHOD
6.1 A wide range of concerns have been raised
about the present SCBA method, by academics, policy makers, practitioners
and interest groups. These can be grouped under five broad headings:
6.1.1 Questions concerning how current benefits
are valued (e.g. travel time savings, where the same unit rates
are used for small and large time savings, and for time savings
and time losses);
6.1.2 Impacts that are not captured by the current
method, such as the severance impacts of road traffic;
6.1.3 Forecasts - both of behaviour change and
of costs and the monetary value of future benefits - are very
difficult to assess and carry large margins of uncertainty that
are not fully reflected in the appraisal;
6.1.4 The method is not consistent with the setting
of objectives and targets, which is how Local Transport plans
are produced; and
6.1.5 Evidence that political decisions do not
mirror the outputs from SCBA (i.e. the benefit/cost ratio).
6.2 In relation to (5), the limited evidence
base which does exist in Norway, Sweden and the UK suggests that
factors other than the benefit/cost ratio are more important in
decisions taken by politicians and that the benefit/cost ratios
might be more influential as a filter to remove poor projects
at an early stage and to appropriately rescope other proposals
to achieve a stronger benefit cost ratio.[60]
7. THE ANSWERS?
7.1 The UK has a well developed methodology for
cost-benefit analysis as part of a broader appraisal method. Considerable
skills and expertise exists within government and consultancies
which allows the procedures to be applied. The government has
invested heavily in providing guidance on how to develop such
proposals. Any change from the current system should be able to
demonstrate that it would lead to more effective outcomes - we
should not "throw out the baby with the bathwater".
Though we note that other countries such as Germany use multi-criteria
analysis to rank investment projects and so alternative approaches
do exist.
7.2 We suggest that the answers lie in a twin-track
approach: In the short term, make good the major weaknesses within
the existing SCBA approach, and In the longer term, take a more
fundamental look at appraisal: what is it for? In what context
is it being applied? What kinds of outputs are most useful and
defensible?
7.3 The Short-term
7.3.1 The treatment of CO2 costs
7.3.1.1 There is a strong debate suggesting the
need to include a monetised estimate for carbon dioxide emissions
which reflects the long-term costs of climate change. This is
one solution to raising the importance of the environment in the
decision-making process, but it seems to be very difficult to
agree on the current costs of CO2 emissions, and forecasts
of future values vary by a factor of ten or more. However, as
noted above, decision-makers have been shown to be able to include
these wider factors without their monetisation. It is not necessarily
the case that monetising carbon will lead to decisions which are
consistent with achieving our carbon targets. It could remain
the case that time savings dominate the appraisal process (as
is the case with High Speed Rail as well as road projects) and
this actually diminishes the importance of carbon in the decision-making
process.
7.3.1.2 The existence of targets for carbon dioxide
reduction also potentially changes the nature of the investment
package proposed. Whilst the government will look for the most
cost-effective reduction strategy across all sectors it is likely
that transport will be asked to make savings which require the
adoption of some carbon reduction policies that do not perform
well in benefit/cost terms (a hypothetical example might be speed
limit reduction). Here a least-cost rather than benefit/cost approach
may be more appropriate.
7.3.2 The treatment of time savings
7.3.2.1 This is one of the most controversial
aspects of existing SCBA applications. Time savings often dominate
other benefits, yet there is growing dissatisfaction with the
current assumptions that all (non-work) travel time savings have
the same unit values. In particular: i. small time savings may
be too small to be perceived by travellers, and have been found
to have smaller unit values empirically; ii. losing time is valued
more highly per unit time than saving time; iii. transport schemes
where time during travel is used productively (e.g. rail) show
smaller benefits from time savings, so investment goes into enhancing
the less the productive modes. Wider concerns have also been expressed.
7.3.2.2 Two issues in particular stand out. First,
that time savings are only short-run gains since daily travel
time budgets have not reduced in response to the cumulative improvements
to the UK transport system over time (Metz), Second, the measured
wider economic benefits are much less than the time savings which
are regarded as proxy values for the former (Wenban-Smith). Enough
data exist to begin to explore the implications of changing the
treatment of time savings within the existing SCBA.
7.3.3 Distributional issues
7.3.3.1 The costs and benefits of schemes vary
both spatially and socially - something which has been paid very
little explicit attention in SCBA methods. The benefit/cost ratio
provides a net estimate of all costs and benefits combined, and
do not highlight the groups or areas that gain or lose from a
proposed scheme. For example, the areas which benefit from time
savings may well be different from those which suffer increased
noise or air pollution resulting, say, from the construction of
a town by-pass, or a high speed rail line. There is considerable
experience of mapping impacts in geography and elsewhere that
could be formally incorporated into SCBA, showing how social groups
and areas are impacted in different ways.
7.3.4 Dealing with contraction
7.3.4.1 Does benefit/cost work in the same way
for projects aimed at infrastructure contraction as it does for
expansion? Clearly, the calculation should be different - the
cost is not one of construction and operation but of savings in
operation and the cost of decommissioning the kit. As Spending
Departments are required, in the Spending Review, to analyse what
they do, what they do not need to do, what they could do better,
then perhaps it is necessary to consider whether the use of the
SCBA approach could usefully contribute to such debates. When
the last significant railway closures ignited under the direction
and enthusiasm of Dr Beeching, the basis on which each line was
closed was based on its immediate individual financial viability.
No account was taken of the effect of closure on the totality
of the railway, nor of the economic or social impacts (CO2
or other) of closure; nor were routes safeguarded for potential
future transport requirements. We might usefully avoid this narrow,
accountancy-based analysis in the current climate.
7.3.5 Un-monetised impacts
7.3.5.1 Transport schemes have significant impacts
which are not monetised, as described above. A comprehensive review
of these is needed, with the flexibility to add more as needs
and values change over time. This review should, therefore, include
a structure for taking account of non-monetised benefits and disbenefits
in the decision making framework.
7.3.6 Decision making: SCBA's role
7.3.6.1 Where does SCBA sit in the decision making
system? A scheme which goes through the "option appraisal
to npv" process described above could still be pointless
if it works against other schemes taking place (e.g. a public
transport system to reduce car use alongside a road widening -
as happened in several cases in the 1980s and 1990s); or if it
doesn't contribute to the town / regional / national objectives;
or if it suffers widespread opposition and protests, even closure,
because of failures of consultation.
7.4 The longer term
7.4.1 Track two would involve taking a much more
fundamental look at appraisal, in two respects. First, by looking
afresh at why appraisals are needed and what purposes they serve
- in the context of the tasks carried out by, and statutory duties
placed on, different bodies (e.g. local highway authorities, Integrated
Transport Authorities, the Highways Agency).
7.4.2 Secondly, by looking at how appraisal is
carried out on different sectors and in different countries. For
example, cost-effectiveness is more widely practised in some sectors
(e.g. where there are firm standards or targets that have to be
met), and some countries prefer a multi-criteria approach to a
benefit/cost approach.
7.4.3 It is always problematic to carry out a
fundamental review of this nature, because of the uncertainties
and implications for schemes in the pipeline or at public enquiry.
But, in this respect the current economic cloud has a silver lining,
in that the current near moratorium on progressing major new transport
schemes presents a once-in-a-lifetime opportunity to carry out
such a fundamental review without prejudicing major transport
investment programmes.
8. THE PROPOSITION
8.1 The Commission provides an independent assessment
of the transport issues of today. Based on the lively debate at
our discussion evening, and parallel debates in the technical
press, it is evident that there is a broadly based unease as to
the fitness-for-purpose of current appraisal methods. This concern
needs to be addressed as a matter of urgency, and the current
restraints on developing major new transport schemes provide an
ideal breathing space to do so. We will also have to take into
account various developments including the impacts of an ageing
population, advances in technology (in transport as well as in
other forms of communication) and the increasing globalisation
of business and society. All will have major impacts on requirements
placed on our transport systems.
8.2 Taking a pragmatic view, we recommend a two-track
approach:
8.3 First, Patches and quick fixes: adding into
the existing framework updates that are do-able within a short
time frame and meet some major concerns. These are likely to include
revisions to the treatment of time savings, better incorporation
of CO2 impacts, and an explicit requirement for SCBA
to be within a decision making framework and not the entire argument.
8.4 Second, A root and branch review: a fundamental
reassessment of what why we carry out appraisals and what we could
learn from the approaches used in other sectors and countries.
NOTE
The Independent Transport Commission was Britain's
first transport and land use think tank and founded in 1999. It
has 10 unpaid Members, a part-time Secretariat, and enjoys links
with the Universities of Southampton and Oxford. The current chairman
is Simon Linnett, vice-chairman of N.M.Rothschild & Sons.
The ITC is a charity funded through private donations. Its chief
sponsors are Go-Ahead Group and Stagecoach.
September 2010
59 Nellthorp, J. and Mackie, P. (2000) The UK Roads
Review-a hedonic model of decision making, Transport Policy,
7, 127-138. Back
60
Eliasson, J. and Lundberg, M. (2010) Do cost-benefit analyses
influence transport investment decisions, Paper presented at the
11th WCTR Conference, Lisbon; Odeck, J. (2010). What determines
decision-maker's preferences for road investments? Evidence from
the Norwegian Road Sector, Transport Reviews, 30(4), 473-494 and
Nellthorp, J. and Mackie, P. (2000). The UK Roads Review-a hedonic
model of decision making, Transport Policy, 7, 127-138. Back
| 
