Evidence suggests that the UK performs poorly on infrastructure compared with some other wealthy countries. This Institute for Government work programme, which is supported by the Project Management Institute, explores how UK economic infrastructure policymaking can be improved.

The aim of this report is to evaluate how the UK appraises infrastructure projects, and in particular the Government’s use of cost benefit analysis. We offer recommendations on how the Government might improve the way it uses cost benefit analysis to accurately include relevant benefits and costs, assess them consistently, and communicate results clearly.

The first publication in this series looked at some of the main flaws in recent and controversial UK ‘megaprojects’.

Subsequent publications will look at:

•          how infrastructure projects are financed, and government’s ability to strike good deals

•          politics and institutions, including -
            *        public consultation,
            *        devolution, and
            *        the role of politicians and experts in decision making.

The UK Government spends tens of billions of pounds every year on economic infrastructure, including energy, transport, water, utilities and digital communication.

Picking the wrong projects risks wasting money now and damaging the economy in the future, but picking the right projects can boost productivity and economic growth over the long term.

Cost benefit analysis (CBA) is a way of summing up the positive and negative impacts of a project. It is one of the key tools the Government uses to decide between competing projects. Yet, as this report demonstrates, it can be misused, inconsistent and poorly communicated. It does not always play a central role in decision making and is too often used to justify decisions that have already been made. As the expensive overruns of the Channel Tunnel and the embarrassing u-turns of the rail electrification programme show, progressing with projects before a thorough analysis has been undertaken can lead to problems.

But despite these issues, CBA is still the best tool available for helping government to appraise and prioritise potential investments. Improving its use is therefore crucial.

In this report, we identify four problems and offer recommendations for overcoming these.

Ministers are interested in projects that can create jobs and increase productivity and Gross Domestic Product (GDP). But existing techniques for estimating ‘dynamic effects’ – those that change the structure of the economy – are -
*        costly to apply,
*        difficult to undertake; and
*        relatively underdeveloped.
Some industry models produce high numbers that lack credibility.

Recommendations

• The Treasury and relevant departments should produce updated guidance on estimating the dynamic effects of infrastructure projects. This should include clear advice on reasonable assumptions and models -
   *    that can be used, and
   *    what kinds of projects merit this broader analysis.

• Ministers and senior civil servants should ensure that claims about dynamic effects are subjected to the same scrutiny as other economic claims, -
  *    either by including them in the economic case;
  *    or by formalising an Infrastructure and Projects Authority review of the strategic case.

• Departments should undertake more research into dynamic effects, including evaluating the dynamic effects of projects after construction.

Cost and time estimates for infrastructure projects are almost always over-optimistic.  This misleads decision makers and can lock ministers into undeliverable targets. It also reduces public trust when these are not met.

• Ministers should be honest about cost uncertainty and potential overruns when making public announcements, particularly for big projects.

• Government must consistently evaluate infrastructure projects. All projects should systematically collect data on -
   *    cost outturns against estimates,
   *    delivery times against estimates,
   *    the size of project teams; and
   *    project length.
Thorough post-project evaluations should be undertaken for all major projects.

• These data should be used to inform appraisals, particularly of cost estimates. The Infrastructure and Projects Authority should collate this information centrally and ensure that it is used by departments to enhance reference class forecasting.

Some costs and benefits, such as impacts on health, safety and the environment, are hard to monetise. Both the public and government care about these impacts, but they are not consistently measured across projects.

• Departments should follow the U.K. Department for Transport and produce standardised guidance on valuation and presentation for their appraisals.

• Departments must consistently incorporate these valuations into decision making.  We recommend a five-step approach, including producing a narrative, benefit–cost ratios and an explanation of whether uncertain effects justify moving a project into a different value-for-money category.

CBA results are not always well understood, by either decision makers or the public.  Information may not be presented transparently or communicated clearly, leading to misunderstandings about critical assumptions and uncertainties.

• Analysts should present and communicate their work more transparently within Whitehall, being clear about critical assumptions, and what might happen if assumptions changed. Departments should learn from the Bank of England and the Office for Budget Responsibility about how to communicate technical information in a way that people will understand.

• Ministers must be fully transparent in communicating to the public the assumptions that underpin CBA results, making the relevant data available to the public and presenting benefit–cost ratios as a range of possible outcomes rather than a single number.

• Increased training and knowledge transfer programmes should be made available to policy, delivery and commercial teams, as well as senior civil servants and ministers, so that they better understand analysts’ work.

• There should be a clearer separation between project promoters and analysts to strengthen the integrity of analysis.

• The Infrastructure and Projects Authority should independently assess CBAs conducted for major projects.

Taken together, these recommendations provide a route map for improving the use of CBA, leading to a better selection of infrastructure projects, fewer cost overruns and, ultimately, a stronger economy.

Cost benefit analysis (CBA) – a way of summing up the positive and negative impacts of a project – is one of the key tools the Government uses to decide between competing projects. CBA is much maligned but it is one of the best tools that governments have to appraise and prioritise investments; most states1 and international organisations2 use a variant of it.

Where the methodology and results of CBA are publicly available, they make explicit the judgements that governments make about the scope and size of project costs and benefits,3 enabling public debate and democratic oversight. In the absence of a tool for systematically appraising investments, deciding which projects go ahead could slip into approving projects with the loudest promoters, or the promoters most willing to make outlandish claims about the low costs and high benefits of their project.

Transport infrastructure tends to have the largest social and environmental impacts, and the greatest likelihood of inducing ‘dynamic changes’* in land use, employment, investment and productivity. As a result, CBA debates have focused on transport projects.

Our analysis is informed by practice across infrastructure sectors and departments, and our recommendations apply to government investment decisions in transport, energy, communications and utilities, across all levels of government.

Future reports will look at:

•         the decision-making process that departments follow when deciding how to finance infrastructure projects

•         how government can best engage with the capital markets to secure private finance for new infrastructure, should it decide that this is the most appropriate option

•         how to reduce national infrastructure policy instability, develop better relationships between central government and various tiers of local government, and increase public engagement in decision making.

A final report will draw together the evidence we have gathered throughout this programme of work to make final recommendations on how to improve infrastructure decision making in the UK. This will be launched at an event in early 2018.

Within Whitehall, CBA is mainly used in three separate stages (see Figure 1):

• First, departments use CBA to value projects.

• Second, the Treasury uses CBA results to help allocate spending under fiscal constraints.

Source: Institute for Government analysis

Both during analysis and at the final decision, the results of CBA, in particular the benefit–cost ratio, will never be the sole criteria for a green light. Deliverability, affordability and the prevailing politics of the day all matter too.

Note: SR = Spending Review. Before 2011, the Department for Transport reported the value for money of schemes approved in Spending Reviews, rather than annually. ‘Very high’ was introduced as a category in April 2009. We report schemes approved before 2011 by Spending Review, noting that these were not eligible for ‘very high’ ratings.

Very high High Medium Low Poor

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2004 SR 2007 SR 2011 2012 2013 2014 2015 2016

As CBA results influence decision making, it follows that CBAs should be as good as they can be. In the box below we set out four criteria for a good CBA.

Better evaluation underpins the first two criteria. Comparing forecasts with actual costs and benefits, and feeding this analysis back into appraisals is crucial for helping departments to produce more accurate estimates.

Sometimes government CBAs meet the four criteria but, as we show in the following chapters, they often fall down on one or more criteria.

Static effects are the direct effects of an investment – in the case of transport, changes that emerge due to a reduction in travel time and costs. This will incorporate some wider economic impacts beyond the individuals and firms using the transport. For example, static effects include increased tax returns from wage and productivity gains created when individuals and firms are brought ‘closer together’ (that is, when the time it takes to travel from place to place is reduced, but firms and consumers do not relocate).

Dynamic effects are the induced effects of an intervention – changes that emerge as a result of capital and labour movement. For example, individuals and firms may decide to relocate in response to a change in transport. This will change economic activity in certain areas, sometimes leading to further productivity gains. Static effects do not include these relocations, and the resulting changes in land use, and so miss part of the economic impact.

* Throughout this report we use the term ‘conventional CBA’ to refer to CBA that includes only static effects.

In addition to accurately incorporating into CBA the genuine dynamic effects of larger schemes, we should also care about ensuring that only relevant benefits are incorporated into the CBAs of smaller schemes. Non-existent dynamic effects should not be used to boost the benefit–cost ratios of the vast majority of projects that are not transformational. Not incorporating dynamic benefits when they are unlikely is as important as including them when they are.

Professor Diane Coyle puts the problem plainly:

Dynamic effects happen over a lengthy period and, by definition, involve structural changes to the economy. Conventional CBA can only analyse the static impacts of projects because the Treasury – and most departments – outline strict conditions as to when investments can be assumed to change the structure of the economy.*

In building models based on the existing economy, conventional CBA will not be able to properly assess impacts in a different future economy, or impacts that would make the economy different in the future. The strategic case for High Speed Two (HS2) explicitly acknowledges this, noting that ‘the benefit–cost ratio methodology was not developed [for schemes] on the scale of HS2’.23

Combined value (£bn)

0 50 100 150 200 250 300 350 400 450 500

Small

(<£100m)

Major

(£100m–£1bn)

Mega

(>£1bn)

Size of scheme

134

schemes

264

schemes

89

schemes

While measuring impact on productivity or GDP can be a useful supplement, it is not complete and cannot substitute for judging the overall value of transport projects, which should incorporate monetised and non-monetised impacts.

There is a case, however, for giving this supplement more weight alongside CBA, if economic growth is the main objective of infrastructure investment.

www.gov.uk/government/consultations/transport-investment-understanding-and-valuing-impacts

To ensure that capturing dynamic effects is valuable, consultants and their clients need to make sure that they are transparent about how they model these effects. If clients do not have access to models and their underlying assumptions, they will not be able to ask constructive questions.36 Without transparency there is a risk that in trying to model dynamic effects,37 imperfect CBAs will be replaced with more complicated black box models.38 This shifts the debate into a much less accessible, technical space, constraining healthy policy debate.

These problems emerge clearly when the touted dynamic effects of a project diverge significantly from conventional CBA results.39

other areas and is therefore 100% displaced, rather than additional. The Department for Transport’s latest

guidance sets out the circumstances in which defaulting from this assumption would be appropriate, however.

system/uploads/attachment_data/file/554783/transport-appraisal-guidance-webtag-consultationdocument.pdf

** Optimism bias is the phenomenon that analysts are likely to make overly optimistic estimates: overestimating

benefits and underestimating costs. How government adjusts for this is explained in the next chapter.

*** Sensitivity testing involves analysts adjusting variables in a model to test their impact on the outcome. This is

used to see how much a certain cost or benefit would need to change for the benefit–cost ratio to change significantly.

Economic growth and rebalancing the economy have been cited as major

effects were not included in the 2013 economic case for HS2. An analysis of the

regional economic impacts of HS2 was therefore commissioned to complement

The size of the benefits estimated by KPMG heightened attention on the

modelling used. The estimates became the focus of public and parliamentary

debates and challenges to HS2. Henry Overman, a former adviser to High Speed

In October 2013, following a Freedom of Information request, critics argued

that there had not been enough focus on the impact of HS2 on those regions

KPMG denies that it misrepresented any figures under pressure from HS2

details of the original report.* But where models are experimental, there needs

to be more scrutiny and a clearer explanation of how estimates were generated,

and their level of uncertainty. Without this, it can easily appear that

transformational effects are being wilfully overvalued, rather than just uncertain.

The HS2 case highlights the need for clear guidance on including dynamic

effects in economic appraisals. There must be a clear method of communicating

uncertainty and different scenarios, including a narrative explanation. It also

points to the importance of transparency and communication in promoting

constructive debate, as we discuss further below.

Guidance on valuing wider impacts has improved significantly over the past 20 years

and the Department for Transport has been at the forefront of the debate.

The 1999 Standing Advisory Committee on Trunk Road Appraisal (SACTRA) report was

its first major consideration of dynamic effects. Its remit was to consider what impact

impacts on the economy should only be included for large projects. In most cases,

dynamic effects are covered by time-savings benefits; to include them would be

supplementary guidance. The 2015 guidance on ‘valuing infrastructure spend’

suggests that ‘supplementary analysis’ of dynamic effects should be presented

The Department for Transport first published guidance on including – static – wider

This, and the greater clarity on dynamic effects (‘indicative monetised impacts’) in the

communication of the way dynamic impacts are expected to play out, and how they

relate to conventional CBA, is positive. This follows through on the recommendations

The Department for Transport has made good progress in attempting to capture all

relevant benefits in the economic case, ensuring that they receive the necessary

scrutiny. It is important now that this is done routinely on projects that claim dynamic

effects. Where other kinds of economic infrastructure projects have transformational

should learn from the Department for Transport’s practice.

Whether Whitehall includes all relevant impacts is only one half of whether CBA

provides useful information. It is also critical that the costs are assessed accurately.

Over-optimistic cost estimates mislead decision makers, locking ministers into

undeliverable targets or unrealistic promises, reducing public faith in government when they are not met.

Cost underestimation refers to the tendency for cost forecasts to be lower than final

costs incurred. The UK is far from alone in suffering from infrastructure cost

underestimation. Internationally, megaproject academic Bent Flyvbjerg estimates that

suggests that railways and tunnels are the transport projects most likely to overrun,

announced early – those announced without a budget commitment – account for a

disproportionate amount of cost overruns. In a dataset of all transport projects valued

at more than AUS$20 million between 2001 and 2015, projects with an early

There is no single explanation for cost underestimation; academics point to three main causes:

look good and push it through the approval process

project; for megaprojects, this can often be because of an (in)ability to assess all

that is required to make a very large project work

from an initial – usually low – estimate (the ‘anchor’ number), and subsequent

tendency not to adjust their estimates far away enough from it.

Our interviewees suggested that most errors are due to a combination of all three,

with particular blame attributed to strategic misrepresentation – usually under

Over-optimistic cost estimates cause significant problems. They damage decision

making and create political problems for ministers; and they make it harder to manage departmental budgets.

First, governments that commit to infrastructure investments based on low early costs

may end up selecting the wrong projects. They can find themselves locked into

projects that no longer look a good bet, as spiralling costs reduce projected value for

by Parliament and the public when this happens, as shown by recent rail electrification projects (see the box below).

Second, misleading analysis leads to difficulties in managing departmental budgets.

The Treasury cannot make appropriate capital funding allocations during Spending

Reviews if it cannot rely on the cost estimates provided by departments. There is also

a danger that cost underestimations make infrastructure projects look better value

2012 were serially underestimated, locking successive transport ministers into

undeliverable commitments. Escalating costs on the Great Western

electrification project raised concerns, and led Sir Patrick McLoughlin, the-then

Secretary of State for Transport, to pause it in June 2015.

Between July 2012 and June 2015, the costs of the Great Western

electrification project rose substantially. Due to a multitude of factors,

including immature cost estimates and project management shortcomings,

Network Rail revised its £1.1bn October 2013 cost estimate to £1.8bn in

September 2014. These revised estimates were assessed and approved as

plausible by the Office of Rail and Road. But even these ‘mature’ cost estimates

were over-optimistic. These revisions rose again to £2.8bn when re-planned

The reasons for the increases are multifaceted. Part of the cost rise was due to

weak project management and planning: detailed requirements were not

Network Rail was too optimistic about the scope of cost reductions that could

be achieved using a factory train, an alternative to labour-intensive

construction. Designs were drawn up and equipment was purchased before

ground-level research, requiring costly equipment modifications and redrawn

specifications. Network Rail further underestimated the costs of ‘route

clearance’ – removing obstacles to electrification, and the associated costs of

getting the 1,800 separate consents to perform work on the line.

Since McLoughlin’s June 2015 pause, subsequent transport ministers have

performed further u-turns. Midland Mainline and TransPennine electrification

was paused alongside Great Western electrification in June 2015, before being

restarted in September 2015. Four Great Western projects were delayed in

November 2016. Finally, electrifying the Great Western line west of Cardiff was

cancelled in July 2017, alongside parts of the Midland Mainline and Lake

District railways. At present, the Government has not committed to funding any

infrastructure enhancements in Network Rail’s next five-year funding period,

This whirlwind of promises has confused almost all external observers, and the

cost estimates, and the Office of Rail and Road enough time to scrutinise them,

they could have announced a more realistic programme. Ministers should care

about whether cost estimates are accurate: if they are not, they will only lead to

political trouble further down the line.

Dealing with unrealistic cost estimates is nothing new. Whitehall is aware of the

problem and the Treasury has standardised advice for addressing it. Based on research

initially conducted by engineering and management consultants Mott MacDonald in

0

200

400

600

800

1,000

1,200

Installation of

electrification equipment

Route

clearance

Programme

management

Signals and

communicaions equipment

Lead design

organisation

Risk and

opportunity

Other

Project costs (£bn)

Estimated cost 2014 Additions in 2016 estimated cost

Project appraisers have the tendency to be over optimistic. Explicit adjustments

should therefore be made to the estimates of a project’s costs, benefits and

The guidance stipulates that an uplift within a specified range (as shown in Figure 5)

be made to forecast costs at the outline business case stage of projects (the second

depends on the amount of risk that has been mitigated. This new figure is used to

confirm whether the economic case for a project would still be robust if historically

observed cost overruns were repeated.

This guidance was a welcome step forward. It aligns with the principles of reference

forecast project costs and benefits using data from similar past projects, albeit with rather blunt categories.

However, there are ongoing questions about how far this guidance is applied in

More broadly, it is unclear whether the guidance has improved the accuracy of cost

estimates in government. There is no publicly accessible database of cost estimates at

different project stages, although departments and agencies responsible for

infrastructure projects should hold such data. As a result, it is not clear to external

observers whether the data are being aggregated and used to inform cost estimates.

Within government departments, there is evidence that more specific data are being

* See, for example, the debate between Heathrow Ltd and the Airports Commission over applying optimism bias

to the rail infrastructure costs associated with a third runway. See Pickard J (2016) ‘Heathrow runway “faces

0a22-11e6-b0f1-61f222853ff3

0%

50%

100%

150%

200%

Suggested optimism bias uplift

Standard

buildings

Non-standard

buildings

Standard civil

engineering

Non-standard

civil engineering

Equipment/

development

Higher adjustment Lower adjustment

precise reference categories: roads, light rail, conventional rail, ‘fixed links’ (bridges

Similarly, the Department for Environment, Food and Rural Affairs (Defra) has a lower

little publicly available evidence in either case that more precise guidance has led to more realistic cost estimates.

More promisingly, there is evidence from Highways England that cost estimates for

POPE has improved appraisal through the collation and use of evaluation data.

0

5

10

15

20

25

30

35

40

Average error in major

project cost forecasts (%)

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Systematic evaluation with data feeding back into appraisal may not be feasible for all projects. But it should be the norm for major projects that we undertake regularly – such as roads, rail enhancements, mid-size utilities investments and flood defences.

Megaprojects will require a different approach. Extra care should be taken not to commit too early, and to consult with operational and delivery officials before announcements. As megaprojects often have few directly comparable predecessors in the UK, reference class forecasting based on past UK projects is unlikely to be as useful. But the general principles of reference class forecasting should still apply. Cost estimates for high-speed rail or major nuclear power plants could be stress-tested by looking at appraisal accuracy in similar international projects.

For some costs and benefits, such as operating costs and journey-time savings, departments endorse a standard way of valuing, monetising and including them in a CBA. Others, however, are harder to value, monetise and include. It is not obvious, for example, how to monetise impacts on health, safety and the environment.

For impacts that are difficult to put a monetary value on, technical improvements will only go so far. It is important that government analysts value these impacts consistently and departments draw on the expertise of others. Encouraging consistency with more guidance will help to ensure that the Treasury is able to compare projects fairly when choosing which to proceed with. It should also help to avoid disputes about valuation methods, which can obscure more fundamental challenges.

Some analysts have suggested that multi-criteria analysis could be a better

appraisal tool than CBA, given the struggle CBA faces to consistently include

analysts assigning numerical values to all project criteria, explicitly weighting

each element of the criteria, and then aggregating them to produce a final set of numbers to rank projects against one another.*

Unlike CBA, multi-criteria analysis is presented as a decision-making rather

than decision-support tool. Current UK guidance is that multi-criteria analysis

may be used alongside CBA where costs and benefits cannot be valued in

Multi-criteria analysis is said to be more transparent than CBA because it forces

decision makers to put an explicit weight on the project criteria that matter to

them. If the numbers behind these weights are published, this should make decision makers more accountable.

However, the theoretical benefits of multi-criteria appraisal are undermined in

practice. It is prone to the same errors as CBA – uncertain data and biased use of

appraisal models – but its greater complexity makes it easier to manipulate. It

can make comparing projects difficult, because the weighting given to various

factors is not always consistent, changing as government priorities change.

CBA, done well, makes comparison possible because the values for different benefits can be standardised.

The five-case business case model that the UK Government now endorses

means that the economic case (the CBA) is now only one of five parts of the

overall business case. As stressed to us by interviewees, decisions are not

As such, the UK is effectively already using a form of multi-criteria analysis,

albeit without the explicit weights required for a full analysis of this kind. This

is preferable as ‘implicit weighting’ requires ministers to outline a clear narrative for their decisions.

Despite extensive guidance, it remains inherently difficult to monetise many of the

impacts of an infrastructure project. For example, valuing loss of life or the loss of

access to a nearby park will always be subjective. Monetising these impacts ‘can work

as long as everyone agrees on the measures used [but] if there is disagreement, proxy

* It is possible to undertake a multi-criteria analysis without aggregating the numbers (partial multi-criteria

analysis) and without explicitly weighting the different options. The results of a government ‘five-case model’

for business cases look similar to an unaggregated multi-criteria analysis that has not been weighted.

measures of value are very easy to pour scorn onto. CBA and any attempt to monetise

But CBA should do its best to include all relevant impacts. That some impacts are

difficult to monetise should not be a reason to neglect them, especially when decision

about what is, and is not, included in a CBA, some consistent numbers would be better than no, or inconsistent, numbers.

The main problem that analysts currently face is the lack of an agreed technique for

valuing these impacts. This can, quite unintentionally, lead to inconsistency. As noted

given variations in size of project and resources available, this flexibility can cause

problems. In the past, it has led to inconsistency in how different departments include

Inconsistency makes comparing projects more difficult, and creates room for

unnecessary valuation disputes, which obscure more fundamental areas of

In addition to the above, we heard that the complexity and presentation of the

Government’s guidance can be challenging for analysts and policymakers to master.

The civil service has done a lot of work to promote appraisal consistency, mainly

The work of Defra and the Department for Transport has been particularly valuable.

which impacts are relevant and need to be assessed; the Department for Transport

poses similar questions in its guidance for practitioners on deciding which economic

Using standardised data is key to comparability. Defra is developing a new dataset for

environmental impacts as part of its work with the Natural Capital Committee, an

National Statistics. This entails attempting to value natural assets: ‘soil, air, water and

market and non-market values of benefits from the UK’s natural capital. Non-market

examples include the value of natural capital as a source of recreation and air

which includes data to value the impacts referred to in appraisal guidance:

journey‑time savings, operating costs, carbon emissions, noise and safety

improvements. The department has also developed a public Wider Impacts Dataset,

Incorporating all relevant benefits, being realistic about costs and valuing impacts

consistently should improve the accuracy of CBAs. But more should be done to

improve the way CBA results are used in decision making.

This requires the results to be well understood: not only by analysts, but also by the

senior civil servants and ministers who use the results. This requires analysts to

present their work transparently and communicate it clearly to policymakers and

ministers. Decision makers must understand the critical assumptions and

uncertainties in CBA, and what CBA does, and does not, explain.

Beyond Whitehall, ministers must improve how CBA’s role in decision making is

communicated to ensure that they do not mislead the public when explaining why particular projects have been selected.

From our interviews, we identified a series of problems stemming from a mixed

understanding of CBA on the part of policymakers, and inadequate communication on the part of analysts.

It is not clear that policymakers always properly understand the purpose of economic

appraisal: to provide an estimate of the net impact of projects, allowing them to be

compared consistently. We heard from analysts that policymakers too often see CBA

dismiss it if analysts are unable to provide a certain, monetisable, figure for all

impacts. If policymakers view CBA as a hurdle to jump, they can put unhelpful

pressure on analysts to find favourable results.

We heard from departmental analysts that they sometimes feel pressured to find

favourable benefit–cost ratios for a minister’s or senior civil servant’s favoured

projects. They told us that this pressure is frequently compounded by additional

pressure – from the same sources – to turn the analysis around quickly and with a high

ensuring analytical rigour, and being seen to ‘get things done’ and contribute to

appraise projects and by the Treasury to prioritise spending, it is important that they

are robust, and that analysts do not feel pressured to find particular results.

Seeing CBA as a hurdle to jump can lead to poor decision making. We were told that it

is not unusual, particularly in larger projects, for CBA to be used to justify a project

that has already been selected for political reasons. One former senior minister told us

are legitimate political reasons for proceeding with a project despite a relatively low

benefit–cost ratio – a government may, for example, wish to prioritise investment in a

particular region. It would, however, be preferable if ministers were open about their

reasoning, rather than attempting to hide behind a retrofitted CBA.

Policymakers should not dismiss the results of CBA that do not give them the

outcomes they want. One senior interviewee responsible for a major project

expressed cynicism about whether CBA could really tell decision makers about the

impact of projects. Decision makers would, in their words, be “kidding [themselves]

that it [CBA] will be a true representation of the value that will be delivered by a

accurately capture. But it suggests a wholesale rejection of the value of CBA which is unwarranted and unhelpful.

This could be mitigated if policymakers better understood the uncertainty in CBA and

long-term forecasts. As we discussed in our first report in this series on infrastructure

Their understanding will be hindered if CBA results are presented uncritically.

Better understanding requires better presentation. Results should not simply be a

single number: this kind of presentation implies a certainty that is unjustified. It also

makes it harder for policymakers to challenge critical assumptions and see how results

sensitivity and scenario analyses should generally be included in presentations

and summary reports to decision makers, rather than just single point estimates…

there are ranges of potential outcomes, and hence [decision makers need] to judge

Most departments have their own guidance on how to present uncertainty. The

appraisal summary table, showing all key impacts, quantitatively and qualitatively

assessed, monetised where relevant, alongside distributional effects.

for sensitivity analyses. The strategic cases for the different HS2 phases clearly

costs and benefits were less clear because they were presented as single figures in different tables.

CBA results are normally presented in tables rather than as graphs or charts.

Departments could learn from other organisations such as the Bank of England and

the Office for Budget Responsibility, which present results in more visually appealing

CBA IS NOT COMMUNICATED CLEARLY AND TRANSPARENTLY 29

ways (see the examples below). The overarching aim of presentation should be to

make CBA as accessible as possible to those without an economics background. The

anyone can use to see the impact of altering demand- and supply-side policies –

should be considered the benchmark for presenting highly technical information in an

number that only analysts understand, it will alienate decision makers and ultimately make CBA less useful, and less used, in Whitehall.

These two charts compare median forecasts with other possible forecasts. In

the first chart, each lighter shade of the fan represents a 10% lower probability

of GDP being within that band. This illustrates increasing uncertainty over time.

The second chart shows the level of uncertainty for one particular forecast, for 2016–17.

-6

-4

-2

0

2

4

6

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Percentage change on a year earlier

March 2012 EFO central forecast

0

0.1

0.2

-1 0 1 2 3 4 5

Percentage change on a year earlier

Probability density

March 2012 EFO central forecast

In these two tables, PVB = present value of benefits, PVC = present value of

costs, WEBs = wider economic benefits, R = route, WSL = Western Southern Link and ESL = Eastern Southern Link.

Here, the ‘most likely’ benefit–cost ratios for different options are listed in one

table, and a separate table outlines the more cautious P90 benefit–cost ratios

(P90 indicates 90% confidence that the benefit–cost ratio will be above this

figure). Probability projection charts could visualise the likelihood of different

options achieving particular benefit–cost ratios but none are provided.

The above charts and tables have been reproduced with permission.

Outside of Whitehall, there is miscommunication and misunderstanding about the role

that CBA plays in decision making. Our interviews with civil servants and ministers –

ministers tend to place most emphasis on the strategic case and wider political

considerations when making decisions. This is not necessarily wrong – the economic

(1) PVC calculation includes discounted revenues from user charges

PVB (£bn) 2010 present value prices R1 R2

WSL

R2

ESL

R3

WSL

R4

WSL

R4

ESL

(1) PVC calculation includes discounted revenues from user charges

case is only part of the reason to go ahead with a project. But the role that CBA plays is

sometimes misrepresented by ministers. This tends to happen in two ways, both of which are damaging.

First, politicians sometimes publicly overemphasise CBA results when they provide

high benefit–cost ratios. When benefit–cost ratios are high, politicians often find it

easy to reach for the CBA as an argument to persuade voters of the merits of a project.

When communicating with the public, the benefit–cost ratio is an easy-to-understand

metric – it sounds authoritative and persuasive – and a scan of press releases and

speeches around major project announcements reveals a tendency for politicians to emphasise it.

But the quantitative certainty it provides is illusory, and focusing on a single

benefit‑cost ratio contradicts Treasury guidance, which states that ‘the output of the

certainty of CBA results can also limit political space for change if projects do not

proceed as planned, and the costs and benefits start to look different from those

announced at the beginning of a project, as successive transport ministers found with rail electrification.

Second, when benefit–cost ratios are low, the principles of CBA are often rejected

wholesale. It is frequently implied that the UK would underinvest in infrastructure if

we relied on conventional CBA. Lord Snape, the former Labour MP, has argued that:

If it were left to the Treasury, we would not have built the M25, the Jubilee line, the

Docklands Light Railway and various other schemes that most people would agree

are essential… [there are few projects that] would ever have passed the

Likewise, in a debate over the East Anglia Rail Franchise, Claire Perry, the former

Parliamentary Under Secretary of State for Transport and now Minister of State for

Climate Change and Industry, argued that:

[E]very major infrastructure project in this country has failed the economic value

test that the Government have imposed on it, because such projects are looked at

Both approaches – either using CBA as a full explanation or rejecting it as a decision

support tool – do more to confuse than to inform the public about the reasons for

going ahead with specific projects. If the public have been misinformed about the

reasons for going ahead with projects, it is much harder for them to make constructive

challenges. It also does a disservice to CBA. By failing to understand, or failing to

communicate, that CBA is not the sole basis for decisions, ministers encourage undue

scepticism towards CBA, both within and beyond Whitehall.

The civil service has recognised the sometimes-poor communication and understanding of CBA on the part of analysts, policymakers and ministers. To date, it has focused on two kinds of reform to mitigate these problems: greater scrutiny and building understanding. Although these have not been specific to infrastructure, they could be replicated to improve the use of CBA for infrastructure projects.

The independent Regulatory Policy Committee (RPC) highlights how greater scrutiny

can incentivise policymakers to take CBA seriously, reducing unwarranted pressures

they may place on analysts. Established in 2009, the RPC scrutinises the regulatory

impact assessments written by departments – the business cases for regulatory

have led to more rigorous analysis. Sense about Science, the Institute for Government

and the Alliance for Useful Evidence found that undertaking impact assessments for

our interviewees who had prepared impact assessments noted that the fear of

receiving a Red or Amber rating on an RPC evaluation motivated analysts and

Separating project promoters and analysts can also strengthen the integrity of

analysis. Both the Department for Transport and Transport for London have separate

teams with responsibility for reviewing appraisals. Such an approach could be adopted more widely.

Attempts to build understanding of CBA beyond analysts have been less formal, but

just as important to ensure that those outside the analytical community understand

and can make constructive challenges to CBA.

The Treasury has led central efforts to build civil servants’ understanding of business

cases by providing training courses for civil servants across the professions, and

drawing up new guidance on writing business cases. The latter culminated in an

For project senior responsible officers, the Infrastructure and Projects Authority

co-ordinates the Major Projects Leadership Academy, which has helped civil servants

to develop skills to manage projects effectively. It would be beneficial if scrutinising

to ensure that senior responsible officers can ask analysts constructive questions.

At the same time, there have been bottom-up efforts to improve relationships through

knowledge transfer. Transport for London, for example, runs training sessions for staff

in policy, delivery and commercial teams on how to create effective business cases.

Project managers and senior leadership are also invited to attend, helping to spread

CBA is a valuable tool in the Government’s arsenal and the best tool available for

helping government to appraise and prioritise potential investments. Getting it right is therefore crucial.

Our recommendations in this chapter focus on how government can improve CBA’s

contribution to appraisal, and deciding which projects to greenlight. We focus on

reforms that will help government to meet our criteria for good CBA – that it includes

all relevant benefits and is realistic about costs, consistent, transparent and well communicated.

Analysts must advance the way they model the impact of transformational projects.

Ministers are interested in projects that can improve productivity, jobs and GDP. Yet,

analysts struggle to analyse these dynamic effects. This can complicate project

prioritisation and lead to unnecessary delays.

Techniques for estimating dynamic impacts do exist but they are costly, difficult to

transparent about the assumptions used, and often produce high numbers that lack

credibility. There is a risk of both undervaluation and overvaluation.

In light of this, departmental guidance on how to include dynamic effects must

improve both to ensure that all relevant benefits for transformational projects are

included, and to ensure that analysts do not assert that marginal projects will have

Government is aware of these challenges and has taken positive steps to improve

practice. It is important that these efforts retain momentum and receive the necessary internal resources.

models that can be used.

There needs to be much greater scrutiny and review of claims about dynamic effects, particularly for transport projects, where these effects can be extensive. Claims about dynamic effects can be central to the strategic case for large infrastructure projects, but the strategic case is not generally subjected to the same levels of scrutiny as the economic case. These effects should therefore be included in the economic case wherever possible.

Where these impacts cannot be included in the economic case, government should formalise an Infrastructure and Projects Authority review of the strategic case, drawing on external academic expertise. Promoters and analysts should also include a narrative of what dynamic effects they hope to achieve, how they aim to realise these, and what they will do to increase their likelihood.

In the UK and internationally, cost and time estimates for infrastructure projects are almost always over-optimistic. While there are mitigating steps that analysts can take, over-optimism is to some degree inevitable, particularly when estimates must be turned around quickly for ambitious projects with large scopes. Ministers should appreciate this in their public announcements.

This does not mean delaying announcements until every penny is accounted for. But it does mean that ministers should be more honest about cost uncertainty and potential overruns in their public announcements, particularly for big projects. Being more upfront about the range of cost estimates will create the political space for ministers to make amendments or cancellations at a later stage. It is right that governments should revisit decisions considering the latest developments, and will find it easier to

do so if limitations of early cost estimates are communicated more clearly.

Learning lessons from the past is critical. Previous projects, whether a stunning success or embarrassing failure, can provide invaluable guidance to those assessing costs and making infrastructure decisions today.

Personal experience is a powerful ally. A civil servant or minister is more likely to challenge and attempt to correct an overly positive assessment if they have been caught out by optimism bias in the past. The challenge that Whitehall faces is institutionalising this memory.

Post-project evaluation should become standard. We rightly expect that large infrastructure projects are subject to thorough economic assessments before securing approval. Yet, comparatively little effort is expended checking whether these assessments were accurate. This is not simply a matter of holding yesterday’s decision makers to account, but of ensuring that tomorrow’s decision makers do not make the same mistakes.

As the Government’s centre of expertise for infrastructure, the Infrastructure and Projects Authority should ensure that thorough post-project evaluations are undertaken for all projects that have appeared on the Government Major Projects Portfolio. All should include an assessment of the extent to which the project achieved its objectives. All projects, of whatever size, should systematically collect data on cost outturns against estimates, delivery times against estimates, the size of project teams and project length.

Data from infrastructure project evaluations should be used to inform appraisals, especially cost estimates. Departments should use this information, collated centrally by the Infrastructure and Projects Authority, to update their optimism bias guidance and enhance reference class forecasting. The example of Highways England is instructive.

Such an approach will be less applicable to big, one-off projects. For these megaprojects, reference class forecasting from past UK projects may not be possible.

But the general principles should still apply, using examples from other countries.

There is no right way to monetise many impacts, but there should be a consistent way.

Greater consistency across departments would help the Treasury to prioritise between projects, particularly where the choice is between completely different options, for example whether to prioritise a transport project or an energy project.

We recommend that other departments follow the Department for Transport in producing standardised guidance on valuation and presentation for their appraisals.

This should build on the expertise that already exists within government, for example Defra’s expertise on the environment. Other parts of the public sector should draw on this knowledge and data where relevant.

To ensure that impacts are valued consistently, government needs to make sure that this guidance is followed. Analysts should be including all relevant impacts based on the latest data. Where impacts are not deemed significant enough to be quantified or detailed, the reasons for this should be explained.

We recommend that the Government takes the following five-stage approach to incorporating the valuations of impacts into the decision-making process.

1. Calculate a benefit–cost ratio that only includes monetary impacts, such as train fares. This would show the strictly commercial return on investment, and the payback period.

2. Calculate a benefit–cost ratio that also includes those impacts that can be straightforwardly and robustly monetised, such as journey-time savings.

3. Show the effect of including impacts where monetisation is possible, but contested – such as environmental impacts – as a sensitivity test.

4. If relevant, show the impact of including uncertain but important dynamic effects as a sensitivity test.

It is essential that the new guidance is understood by policymakers. Its rollout should therefore be accompanied by an extensive training programme with appropriate support provided for senior civil servants and ministers, as well as analysts.

Our research found that policymakers and ministers do not always seem to have a good grip on CBA, viewing it as either a hurdle to jump or a decision-making tool.  Some interviewees had concerns that uncertainties and assumptions were not fully understood at the senior level.

Ministers must improve the way they communicate CBA to the public. This means being transparent about the assumptions that underpin CBAs, making the relevant data available to the public and presenting benefit–cost ratios as a range of possible outcomes rather than a single number. A better understanding of CBA within Whitehall should facilitate better ministerial communication with the public.

Furthermore, greater honesty with the public about the uncertainty of costs, timescales and likely benefits when projects are announced would provide more political space for ministers to make amendments or cancellations at a later stage. The Government should revisit decisions considering new technological, social or economic developments and will find it easier to do so if the assumptions underpinning CBAs are communicated more clearly.

Training for analysts has improved over time and is generally of a high quality. But it is critical that training is not limited to analysts. Senior civil servants and ministers, the people making decisions, need to understand the sensitivity of estimates and have the confidence to challenge underpinning assumptions. The establishment of the Major Project Leadership Academy is a positive step forward for developing the skills of senior responsible officers and we would like to see the Academy expand knowledge transfer programmes for project leaders. The programme of activities for the Academy alumni group should be expanded. The skills of ministers are rarely nurtured in a formal way. We recommend that ministers with infrastructure responsibility should have deeper and more regular engagement with the Academy.

Decision making is improved if officials in policy, delivery and commercial teams have a better understanding of analysts’ work. At Transport for London, these staff are invited to attend sessions on forming business plans that explain the creation of the economic case. Similar training should be made available more widely. The Infrastructure and Projects Authority is well placed to provide this, with priority given to civil servants from departments such as the Department for Transport and the Department for Business, Energy and Industrial Strategy, which undertake significant economic modelling for infrastructure projects.

The Infrastructure and Projects Authority already co-ordinates Transforming Together, a community of practice that brings together staff working on transformation projects to share their work and develop a joint approach to delivering transformation. We recommend that an equivalent is created for infrastructure, potentially in partnership with Exploring Economics. Such a network would enable analysts and others to discuss best practice, including the latest modelling tools and evidence.

Creating a clearer separation between project promoters and analysts can strengthen the integrity of analysis. We recommend that government departments and agencies not currently set up in this way should follow the example of the Department for Transport and Transport for London, which have separate teams with responsibility for reviewing appraisals. A balance must be struck between institutional separation and joint working. Good economic analysis will be an iterative process between analysts and policymakers, but it is important that CBAs are not just post-hoc justifications of decisions already made.

imperial.ac.uk/ref/Public/UoA%2014%20-%20Civil%20and%20Construction%20

Engineering/Wider%20economic%20Impacts/K%20Worsley%20report.pdf

2 Sartori D, Catalano G, Genco M, Pancotti C, Sirtori E, Vignetti S and Del Bo C (2014)

http://ec.europa.eu/regional_policy/sources/docgener/studies/pdf/cba_guide.pdf

php/2016/10/a-land-built-by-economists

Infrastructure%20report%20%28final%29r.pdf

downloadables/Transport_policy_appraisal_decision_making_worsley_mackie_May_2015_final_report.pdf

data/file/220541/green_book_complete.pdf

government/uploads/system/uploads/attachment_data/file/386126/TIEP_Report.pdf

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/554783/transport-appraisal-guidance-webtag-consultation-document.pdf

updated 2017, retrieved 11 August 2017,

www.gov.uk/guidance/transport-analysis-guidance-webtag

https://workspace.imperial.ac.uk/ref/Public/UoA%2014%20-%20Civil%20and%20Construction%20Engineering/

Wider%20economic%20Impacts/K%20Worsley%20report.pdf

attachment_data/file/567908/Treasury_approvals_process_guidance_final.pdf

files/publications/Infrastructure%20report%20%28final%29r.pdf

data/file/567908/Treasury_approvals_process_guidance_final.pdf

www.gov.uk/government/uploads/system/uploads/attachment_data/file/469317/

green_book_guidance_public_sector_business_cases_2015_update.pdf

percentage-of-dft-s-appraised-project-spending-that-is-assessed-as-good-or-verygood-

value-for-money

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/554783/transport-appraisal-guidance-webtag-consultation-document.pdf

transformational-infrastructure-North_Aug2014.pdf

http://webarchive.nationalarchives.gov.uk/20081230093524/http:/www.dft.gov.uk/

about/strategy/transportstrategy/eddingtonstudy/

government/publications/national-infrastructure-and-construction-pipeline-2016

wp-content/uploads/2017/07/Laying-the-Foundations-the-Industrial-Strategy-Commission.pdf

22 Coyle D (2016) ‘Infrastructure cannot be built on empty purses and big promises’,

www.ft.com/content/4bb57c70-e9cd-11e5-888e-2eadd5fbc4a4

attachment_data/file/260525/strategic-case.pdf

www.nzta.govt.nz/assets/resources/research/reports/566/docs/566.pdf

25 Laird JJ, Nash C and Mackie PJ (2014) ‘Transformational transport infrastructure: cost

26 Institute for Government, Infrastructure Project interview, 2017.

27 Stewart J (2013) ‘Give the economic benefits of infrastructure a chance to breathe’,

network/2013/feb/11/economic-benefits-infrastructure-greater-manchester

28 Greater Manchester Combined Authority, Association of Greater

Manchester Authorities, Greater Manchester Local Enterprise Partnership

http://www.tfgm.com/ltp3/Documents/GMGRP-Transport-Strategy.pdf

House of Commons Treasury Committee, p. 32,

http://data.parliament.uk/writtenevidence/committeeevidence.svc/

evidencedocument/treasury-committee/autumn-statement-2016/oral/44537.pdf

30 Institute for Government, Infrastructure Project interview, 2017.

www.gov.uk/government/publications/autumn-statement-2016-documents/

autumn-statement-2016

32 Lowe J (2014) ‘Developing better business cases using the five case model; for better

decision making and better outcomes’, APMG International webinar, 4 September,

using-the-five-case-model-for-better-decision-making-and-better-outcomes

attachment_data/file/630704/value-for-money-framework.pdf

34 Dobes L and Leung J (2015) ‘Wider economic impacts in transport infrastructure cost

http://press-files.anu.edu.au/downloads/press/p332753/html/argument02.

xhtml?referer=1750&page=11#

35 Institute for Government, Infrastructure Project interview, 2017.

36 Institute for Government, Infrastructure Project interviews, 2017.

37 Dobes L and Leung J (2016) ‘Wider economic impacts in transport infrastructure cost

http://press-files.anu.edu.au/downloads/press/p332753/html/argument02.

xhtml?referer=1750&page=11#

38 Rosewell B (2012) ‘21st century policy development’, comment, 21 March, retrieved 21

39 Institute for Government, Infrastructure Project interview, 2017.

Economic%20case_0.pdf

Regional%20Economic%20Impacts.pdf

43 Topham G (2013) ‘KPMG to face MPs again over HS2 report’,

www.theguardian.com/uk-news/2013/nov/11/kpmg-hs2-mps-forecast-report

44 BBC News (2013) ‘HS2 “losers” revealed as report shows potential impact’, BBC News,

committees-a-z/commons-select/treasury-committee/inquiries1/parliament-2010/

economics-of-hs2

uk/20090224130217/http://www.dft.gov.uk/pgr/economics/sactra/

transportandtheeconomyfullre3148

data/file/417822/PU1798_Valuing_Infrastructure_Spend_-_lastest_draft.pdf

nationalarchives.gov.uk/+/http:/www.dft.gov.uk/pgr/economics/rdg/webia/

webmethodology/sportwidereconomicbenefi3137.pdf

www.gov.uk/government/consultations/transport-investment-understanding-andvaluing-impacts

uploads/attachment_data/file/627490/value-for-money-supplementary-guidanceon-

categories.pdf

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/386126/TIEP_Report.pdf

52 HM Treasury, Infrastructure and Projects Authority and Jones A (2017) ‘Billion pound

connectivity boost to make buffering a thing of the past’, news story, GOV.UK, 3 July,

to-make-buffering-a-thing-of-the-past

53 Flyvbjerg B (2014) ‘What you should know about megaprojects, and why: an overview’,

https://arxiv.org/ftp/arxiv/papers/1409/1409.0003.pdf

https://grattan.edu.au/report/cost-overruns-in-transport-infrastructure

56 Flyvbjerg B, Garbuio M and Lovallo D (2009) ‘Delusion and deception in large

infrastructure projects: two models for explaining and preventing executive disaster’,

in-government-projects.pdf

59 Institute for Government, Infrastructure Project interviews, 2017.

60 Brown R, Ashley R and Farrelly M (2011) ‘Political and professional agency entrapment:

pp. 4037–50.

https://grattan.edu.au/report/cost-overruns-in-transport-infrastructure

http://researchbriefings.parliament.uk/ResearchBriefing/Summary/SN05907

uploads/attachment_data/file/630674/high-level-output-specification-web.pdf

66 BBC News (2017) ‘Crossrail 2: support by government “outrageous” after northern

snub’, BBC News, 25 July, retrieved 8 August 2017,

www.bbc.co.uk/news/uk-england-40708531

67 Welsh Government (2017) ‘Ken Skates demands £700m from UK Government for

scrapping electrification across South Wales’, news release, 21 July 2017, retrieved 1

68 Burnham A (2017) Letter to Chris Grayling, 22 July, retrieved 1 September 2017,

https://drive.google.com/file/d/0BxfytYfkWADUVjE4eDVIZS1FR1U/view

guidance-optimism-bias

publications/Infrastructure%20report%20%28final%29r.pdf

72 Flyvbjerg B (2013) ‘Quality control and due diligence in project management: getting

vol. 31, no. 5, pp. 760–74.

attachment_data/file/625380/TAG_unit_a1.2_cost_estimation_jul17.pdf

Department for Transport, retrieved 4 September 2017,

www.gov.uk/government/publications/webtag-forthcoming-changes

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/576976/dft-optimism-bias-study.pdf

www.gov.uk/government/uploads/system/uploads/attachment_data/file/181443/ fcd3update0303.pdf

uploads/attachment_data/file/497240/POPE___Meta_2015___summary___final.pdf

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/497241/POPE___Meta_2015_Final_210116_-_FINAL.pdf

79 Institute for Government, Infrastructure Project interviews, 2017.

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/497241/POPE___Meta_2015_Final_210116_-_FINAL.pdf

81 Topham G and Elgot J (2017) ‘Heathrow night flights to continue until third runway is

reaction-to-the-governments-hs2-project-8478614.html

www.pc.gov.au/research/supporting/non-market-valuation/non-market-valuation.pdf

the-green-book-appraisal-and-evaluation-in-central-governent

91 Annema JA, Mouter N and Razaei J (2015) ‘Cost–benefit analysis (CBA), or multi-criteria

decision-making (MCDM) or both: politicians’ perspective in transport policy appraisal’,

www.sciencedirect.com/science/article/pii/S2352146515002197

92 Department for Communities and Local Government (2009) Multi-Criteria Analyis: A

manual, Department for Communities and Local Government, p. 5, retrieved 11

attachment_data/file/7612/1132618.pdf

93 Australian Bureau of Agricultural and Resource Economics and Sciences (2017)

‘Multi-criteria analysis (MCAS-S)’, Department of Agriculture and Water Resources,

Australian Government, retrieved 7 August 2017,

www.agriculture.gov.au/abares/aclump/multi-criteria-analysis

94 NZ Transport Agency (2017) ‘Multi criteria analysis for transport business cases

nz/resources/17-02-new-multi-criteria-analysis-for-transport-business-casesguidelines-for-consultation/

95 Gühnemann A, Laird JJ and Pearman AD (2012) ‘Combining cost–benefit and multicriteria

analysis to prioritise a national road infrastructure programme’,

www.sciencedirect.com/science/article/pii/S0967070X12000753

96 Institute for Government, Infrastructure Project interviews, 2017.

97 Durrant D (2013) ‘Evaluating grand infrastructure projects like HS2 using cost benefit

analysis may actually make the case harder to make’, blog,

LSE British Politics and Policy, 12 September 2013, retrieved 30 June 2017,

http://blogs.lse.ac.uk/politicsandpolicy/36311/

98 Nellthorp J and Mackie P (2000) ‘The UK roads review—a hedonic model of decision

www.sciencedirect.com/science/article/pii/S0967070X00000020

system/uploads/attachment_data/file/220541/green_book_complete.pdf

National Audit Office, 2011, pp. 15–17,

www.nao.org.uk/wp-content/uploads/2011/05/1012_Option_Appraisal.pdf

101 Flyvbjerg B, Skamris Holm MK and Buhl SL (2004) ‘What causes cost overrun in

https://arxiv.org/ftp/arxiv/papers/1304/1304.4476.pdf

103 Institute for Government, Infrastructure Project interview, 2017.

uploads/attachment_data/file/191500/Accounting_for_enviornomental_impacts.pdf

www.gov.uk/guidance/assessing-environmental-impact-guidance

file/427091/webtag-tag-unit-a2-1-wider-impacts.pdf#nameddest=chptr03

www.gov.uk/government/groups/natural-capital-committee

uksectoraccounts/methodologies/naturalcapital

environmentalaccounts/bulletins/uknaturalcapital/monetaryestimates2016

111 Curnow J (2017) ‘How do you put a value on “natural capital”?’, blog,

https://quarterly.blog.gov.uk/2017/08/02/how-do-you-put-a-value-on-natural-capital

GOV.UK, retrieved 30 June 2017,

www.gov.uk/government/publications/webtag-economic-impacts-worksheets

114 Institute for Government, Infrastructure Project interviews, 2017.

116 Institute for Government, Infrastructure Project interview, 2017.

Infrastructure%20report%20%28final%29r.pdf

attachment_data/file/220541/green_book_complete.pdf

for Transport, retrieved 10 August 2017,

www.gov.uk/government/publications/webtag-appraisal-tables

file/260525/strategic-case.pdf

attachment_data/file/629393/high-speed-two-phase-two-strategic-case.pdf

consultation/user_uploads/lower-thames-crossing-consultation-summarybusiness-case.pdf

125 Greenway A (2016) ‘Policy without the hot air’, comment, 30 May, retrieved 10 August

uploads/Briefing-paper-No4-How-we-present-uncertainty.pdf

user_uploads/lower-thames-crossing-consultation-summary-businesscase.pdf

128 Mullen CA and Marsden G (2015) ‘Transport, economic competitiveness and

http://eprints.whiterose.ac.uk/90360/2/Transport_and_comp_text_Aug%2015_

final%20%281%29.pdf

129 Annema JA, Mouter N and Razei J (2015) ‘Cost benefit analysis (CBA), or multi-criteria

decision-making (MCDM) or both: politicians’ perspective in transport policy appraisal’,

www.sciencedirect.com/science/article/pii/S2352146515002197

file/469317/green_book_guidance_public_sector_business_cases_2015_update.pdf

131 Lord Snape, cited in Lords Hansard (2013) ‘High Speed Rail (Preparation) Bill:

second reading (and remaining stages)’, retrieved 4 September 2017,

www.publications.parliament.uk/pa/ld201314/ldhansrd/text/131119-0003.htm

132 Perry C (2015) cited in House of Commons Debate, 16 December 2015, c1665,

53.0&s=Jubilee+Line+Extension#g1665.0

133 Regulatory Policy Committee (2014) ‘Regulatory Policy Committee: recommendations

used when scrutinising impact assessments’, GOV.UK, retrieved 27 June 2016,

www.gov.uk/government/publications/how-the-regulatory-policy-committeescrutinises-

impact-assessments/regulatory-policy-committee-recommendationsused-

when-scrutinising-impact-assessments

134 Sense About Science, Institute for Government and Alliance for Useful Evidence (2016)

uploads/2016/11/SaS-Transparency-of-Evidence-2016-Nov.pdf

135 Institute for Government, Infrastructure Project interviews, 2017.

www.gov.uk/government/uploads/system/uploads/attachment_data/file/469317/

green_book_guidance_public_sector_business_cases_2015_update.pdf

www.gov.uk/government/uploads/system/uploads/attachment_data/file/405600/

MPLA_Handbook_for_gov_uk.pdf

138 Institute for Government, Infrastructure Project interview, 2017.

139 Vickerman RW (2008) ‘Cost–benefit analysis and the wider economic benefits from

141 Department for Transport (2016) ‘Understanding and valuing impacts of transport

investment: updating wider economic impacts guidance’, Department for Transport,

understanding-and-valuing-impacts

The Stationery Office.

143 Innovate UK and Department for Business, Energy and Industrial Strategy (2017)

‘Industrial Strategy Challenge Fund: joint research and innovation’,

Innovate UK and Department for Business, Energy and Industrial Strategy,

https://www.gov.uk/government/collections/industrial-strategy-challenge-fund-jointresearch-and-innovation

attachment_data/file/576976/dft-optimism-bias-study.pdf

www.gov.uk/government/uploads/system/uploads/attachment_data/file/205209/

Getting_a_grip_Lord_Browne_major_project_review_Mar-2013.pdf

Department for Transport, retrieved 23 August 2017,

www.gov.uk/government/publications/dft-value-for-money-framework

147 Bearpark T, Heron A and Glover B (2017) ‘Economics in government: more open, more

more-diverse-more-influential

Department for Transport, retrieved 23 August 2017,

www.gov.uk/government/publications/dft-value-for-money-framework

149 Atkins G (2017) ‘Garden Bridge raises questions about infrastructure decisions’,

blog, Institute for Government, 13 February, retrieved 16 June 2017,

www.instituteforgovernment.org.uk/blog/garden-bridge-raises-questions-aboutinfrastructure-decisions

https://www.gov.uk/government/uploads/system/uploads/attachment_data/

file/205209/Getting_a_grip_Lord_Browne_major_project_review_Mar-2013.pdf

Researcher in November 2016. Prior to this, he worked as a

Researcher at DragonGate, a public sector consultancy,

working on a range of projects related to local government

and the broader public sector; and at a research agency,

Populus, on politics and corporate reputation.

March 2017. He currently leads the Institute’s work on

infrastructure. Nick was previously Public Services Manager

at the National Council for Voluntary Organisations. He has

also worked at Children England, London Youth and as a

parliamentary researcher.

on the infrastructure policymaking project. Prior to joining

the Institute, Tess completed an internship in public

diplomacy at the Australian High Commission in London.

The authors would like to thank colleagues Emma Norris,

Raphael Hogarth, Bronwen Maddox, Jill Rutter and Nicole

Valentinuzzi at the Institute for Government for invaluable

advice throughout this research. Thanks to the

communications team at the Institute for Government for

their support. We are also grateful to those who read early

drafts of this report, particularly Professor Denise Bower,

Professor Henry Overman and Rachel Wolf. We would like

to thank all those who generously agreed to talk to us

throughout our research. Finally, we are very grateful to the

Project Management Institute for supporting this work. All

errors and omissions are, of course, our responsibility alone.

We provide rigorous research and analysis, topical commentary and public events to explore the key challenges facing government.

We offer a space for discussion and fresh thinking, to help senior politicians and civil servants think differently and bring about change.

The Institute for Government is a registered charity in England and Wales (No.1123926) with cross-party

governance. Our main funder is the Gatsby Charitable Foundation, one of the Sainsbury Family Charitable Trusts.