Written evidence from Chris Riley (TE
108)
1. The Committee has invited evidence on five
issues relevant to its inquiry on Transport and the Economy. This
paper considers each in turn.
Have the UK's Economic Conditions Materially changed
since the Eddington Transport Study and, if so, does this affect
the relationship between transport spending and UK Economic Growth?
2. Since the Eddington Study was completed in
2006, estimates of sustainable growth and the productive potential
of the economy have been reduced in light of the recession and
the impact of the financial crisis. Estimates by the OBR in June
suggest that trend GDP is 9% below the estimates made immediately
before the recession, and is likely to grow less rapidly in the
future. This means that both the level and the growth rate of
transport demand will be lower in the medium and longer term than
previously thought. The recession has depressed demand for all
transport modes, and while demand will recover it cannot be expected
to reach the path assumed in 2006.
3. This means that pressure on existing transport
capacity will be less than if the economy had remained on its
previous track, and hence the need for investment in capacity
will be correspondingly reduced. Rates of return and Benefit-Cost
Ratios (BCRs) for transport schemes in all modes will be lower
than otherwise.
4. Looking at causation the other way invites
the question: could more or better transport spending help boost
economic growth and offset some of the recent deterioration in
economic prospects? The answer in principle is yes, given the
kind of economic impacts identified in transport appraisals and
other sources of evidence. A DfT working paper on wider economic
benefits in 2005 set out how appraisal effects translate into
effects on GDP.[113]
5. But Eddington concluded that:
- ¾ in
mature economies like the UK, with well-developed transport networks,
transport improvements can help primarily by relaxing constraints
on growth, for example by reducing congestion, journey times and
other transport costs, increasing the effective size of local
labour markets, removing obstacles to development and agglomeration;
and
- ¾ improving
connectivity by making new destinations available, and thus previously
inaccessible places accessible, could in principle have substantial
effects, but that is less likely to be important in the UK than
in countries with less well developed transport systems.
6. These conclusions still seem valid. Improved
transport is unlikely to boost growth much in areas of the economy
where the underlying growth conditions are not favourable. The
two-way-road argument[114]
means that improved transport can in some circumstances reduce
economic activity in places which become better connected.
7. A key question for policy makers is whether
transport improvements are more effective in boosting growth than
other forms of spending, such as education and skills, lower taxes
etc. The available literature does not provide clear cut answers
to this question.
What type of Transport Spending should be prioritised,
in the context of an overall spending reduction, in order best
to support regional and national economic growth?
8. Eddington argued that the priority for transport
policy in the UK should be improving the performance of existing
transport networks: especially congested urban areas, key inter-urban
corridors and key international gateways. That still seems right.
This is where the constraints on growth are likely to be greatest.
Many commentators stress the key importance of improving urban
transport systems.
9. Eddington also argued that it was vital to
"get prices right". That too is still correct; we should
focus on pricing for roads as well as public transport, and it's
far from clear that prices are currently right. Some road users
currently pay less than the social costs they impose - including
congestion and climate change.
10. If the aim is to make the greatest possible
contribution to economic growth, transport spending should focus
on schemes which offer the greatest economic returns. This would
mean adopting a narrower criterion than the BCR currently used
in appraisal, omitting for example environmental and social impacts.
Ranking transport options by their contribution to the economy
would not correspond to a ranking based on the wider concept of
welfare currently used, though there would be considerable overlap.
11. The version of DfT methodology employed by
Eddington would probably in general have favoured road schemes
over more expensive rail schemes, and smaller schemes over more
speculative grand projects. But the changes introduced in 2009
(notably to the treatment of indirect taxes) would now tip the
balance more towards public transport.
12. Phil Goodwin's evidence to the Committee
discusses this in more detail. His estimates confirm the high
rates of return for small schemes such as cycling and "smarter
choices" - measures to influence travel behaviour, such as
travel plans - but suggest a very different ordering of priorities
between roads and public transport. Returns for roads are now
estimated by Goodwin to be much lower, but the numbers need to
be examined very carefully. Two key assumptions in particular,
and hence the conclusion itself, may be questioned:
- ¾ that
lower road traffic growth causes as large a reduction in the benefits
of road investment as the introduction of road pricing would.
This is not borne out by other studies, including the DfT's Road
Pricing Feasibility Study in 2004;[115]
and
- ¾ that
demand for public transport, and hence investment returns, are
not affected by lower growth in the economy. This too is not borne
out by empirical evidence.
13. Much depends on how well the appraisals of
individual schemes - particularly large infrastructure schemes
- capture the wider benefits associated with land use changes
and dynamic factors which can lead to cumulative changes in the
economy. The formal models used in transport appraisal are not
always very good at dealing with such effects, and is desirable
to use the best available evidence on this from whatever source,
as discussed below.
14. High Speed Rail is evidently a priority for
the current Government, and the case for this rests on three main
planks:
- ¾ rapidly
increasing demand for rail travel, which will put increasing pressure
on existing capacity (notably the West Coast Main Line);
- ¾ the
fact that speed is very important for medium and long distance
journeys, and values of time are high for some travellers who
are therefore prepared to pay more for high speed services; and
- ¾ environmental
benefits, which are likely to be significant if passengers are
attracted in significant numbers from domestic air services.
15. Extending the high speed network up to and
beyond Birmingham could therefore be very worthwhile if appraisal
demonstrates high BCRs by comparison with other schemes. But the
numbers produced by HS2 are not particularly high. The economic
benefits of additional investment in high speed rail could be
substantial, but have yet to be demonstrated.
How should the balance between Revenue and Capital
Expenditure be altered?
16. It is not clear in macro terms how the overall
balance between current and capital spending should be changed.
Ultimately, this depends on assessments of value for money (VfM)
for particular projects. But there are some general points of
principle to be made.
17. First, capital projects with a good case
on VfM grounds could and should be financed by borrowing, not
by current taxpayers, because the benefits accrue to future taxpayers.
This is the "golden rule". Failure to invest in worthwhile
projects reduces future economic growth - it reduces debt, but
also reduces GDP. Investing now, when there are spare resources
in the economy and cost pressures are relatively subdued, makes
very good economic sense. If the projects are demonstrably good
VfM, it is very unlikely that more borrowing for this purpose
would lead to a loss of confidence in financial markets.
18. Second, maintenance of existing transport
assets - for example roads and the rail network - should not be
neglected, because that simply stores up a need for higher capital
spending in the future. Any unwarranted saving in maintenance
spending may help to reduce public borrowing and debt, but it
also leads to a reduction in the value of national assets.
19. Third, existing assets should be used as
efficiently as possible, especially at times of fiscal stringency
when investment in new capacity has to be strictly rationed. That
is clearly a priority for rail, but also for the road network.
20. Road pricing, as championed by Eddington
and others (eg the RAC Foundation), is the most economically efficient
way of achieving this, and has the added advantage that it generates
revenue which can be used to fund capital and current spending
- whether on transport or elsewhere - while reducing the need
for additional investment in road capacity. It is the politics
of road pricing which appears to be the biggest obstacle to progress;
the economic case is clear.
21. There is a read across to rail fares and
subsidy for the rail industry. Current rail subsidies are partly
a reflection of the fact that roads are under-priced - road users
don't meet the full social costs of road travel (in particular
congestion, but also carbon). Simply putting up rail fares to
help fund increases in capacity will discourage rail demand and
push more traffic onto the roads, but with road pricing this would
not be a problem and rail subsidies could be reduced.
22. Finally, there is clear scope for better
targeting of bus subsidy, as a recent Oxera study for the LGA
has argued.[116]
The existing concessionary fares policy is not well targeted on
those in need, and Bus Service Operators Grant (BSOG) is not well
designed to address local conditions and the environment. Better
targeting would permit funds either to be reallocated more productively
within the bus industry or to local transport more generally,
or savings to be made without damaging it. But the economic case
for well designed bus subsidy is strong (as Phil Goodwin's numbers
suggest).
Are the current methods for assessing proposed
transport schemes satisfactory?
23. The appraisal methods used have improved
in recent years, but further improvements are needed. The degree
of sophistication and the type of modelling used will depend on
the type of scheme or strategy to be appraised and the stage which
the decision process has reached. But the basic methodology is
the same.
24. Some commentators criticise the welfare economics
basis of present methods, but this approach has the great advantage
of covering the full range of impacts on the public in a coherent
and transparent manner. The main problems with implementing the
present approach are empirical, and some impacts are very difficult
to measure. But empirical improvements are being made all the
time, with formal measurement and monetisation being applied to
an increasing range of impacts.
25. Where evidence on welfare impacts is incomplete,
or where particular emphasis is needed on specific kinds of effect,
other analyses can be used to supplement the basic welfare approach
- for example analysis of distributional effects, impacts on economic
growth or the environment, or other criteria. If necessary these
various analyses and other factors can be combined in a more formal
multi-criteria decision analysis (MCDA) approach. But a simple
checklist approach, as some have advocated, or a focus solely
on contributions of transport schemes to government objectives
- which may or may not be sensible - are not adequate substitutes
for rigorous analysis within a sound conceptual framework.
26. Some commentators criticise the important
role of time savings in transport appraisal, but this is misguided.
The benefits of transport arise because of the connectivity they
provide, and as a recent Oxera study has made clear, this essentially
reflects the availability of potential travel destinations and
the time, reliability and cost - or generalised cost - incurred
in reaching them.[117]
In the UK, with its highly developed transport system, the latter
will be most important. Improvements in journey times, reliability
and cost may of course lead to a range of wider effects on the
economy, and these can and should be modelled for inclusion in
the appraisal.
27. Some argue that small time savings in particular
should be ignored, because people don't appear to value them.
But this makes no sense. Small changes may eventually cumulate
to larger changes and cannot be ignored. To do so would introduce
a wholly unjustified distinction between the overall benefits
of a succession of small scale projects, each leading to incremental
changes, and large scale projects delivering the same overall
effect. There are no doubt some improvements that can be made
to the treatment of time savings - for example the valuation of
working time spent on public transport - but the basic approach
is sound. It is, of course, possible to include the distribution
of time savings as a supplementary analysis, as DfT proposes.
28. Some commentators argue that environmental
and some other impacts should not be monetised, because such impacts
- for example on human life, or endangered species - are beyond
price.[118]
There is no evidence that this is so. Society is neither willing
nor able to pay an infinite amount to reduce risks to human life
- or to preserve tigers. Valuation helps to make judgements on
these matters transparent, and should be pursued as far as our
techniques allow. It has already been extended to a range of environmental
and safety impacts, for example. Willingness to pay for reductions
in risk or to mitigate adverse impacts is a sound measure of social
preferences. Where good data are not available, or where it is
not possible to design suitable studies to obtain them, informed
judgements need to be made.
29. The current approach to appraisal takes account
of wider economic benefits of transport, going well beyond time
savings, including agglomeration effects, higher output in imperfectly
competitive markets, and labour market effects. This is an important
step forward.
30. Agglomeration benefits are potentially very
significant, for example, as seen in the Crossrail appraisal.
But as well as effects resulting from lower transport costs, which
are included in the standard approach, it is also very important
to take full account of land use changes and relocation of jobs.
At the moment this doesn't necessarily happen; it is allowed only
as an optional variant on the central assessment. Good land use
transport interaction (LUTI) models are needed to do this properly,
and this approach should be encouraged because wider benefits
may otherwise be considerably understated. Such modelling is not
straightforward, not least because of interaction with the planning
system, but it is an important area for future development. Even
when such modelling of wider benefits is available, however, it
may well be desirable to utilise the results of non model-based
studies to inform assessment of such effects, not all of which
may be susceptible to formal modelling given the current state
of knowledge.
31. Some kinds of wider effect are omitted completely
in the current methodology, such as impacts on competition, overseas
trade and investment. In principle these effects could be important,
and should be analysed. But the evidence base is weak in this
area.
How will schemes be planned in the absence of
regional bodies and following the revocation and abolition of
Regional Spatial Strategies?
32. The nature of transport planning and decision
taking should be different for strategic national networks, like
strategic roads and inter-urban rail systems, and local networks.
The former remain the responsibility of central government, and
the latter will now be the responsibility of local authorities
and/or Local Enterprise Partnerships.
33. Local authority boundaries do not necessarily
relate well to the wider areas served by local transport systems.
London is a very good example, and the mayor has a central role
in regional transport planning. A similar regime could potentially
bring benefits to other city regions, though it is generally accepted
that administrative or statistical regions, as covered by the
RDAs, are not the appropriate concept.
34. In the absence of mayors for city regions,
local authorities need to coordinate their planning of transport
schemes, and the Government should take responsibility for facilitating
this. Whether the new Local Economic Partnerships will fit the
bill remains to be seen. It will depend on the way they are set
up and the powers they have. In principle they could be very effective,
but the scope and coverage of partnerships announced so far do
not appear to obviate the need for further coordination.
35. However local and regional transport planning
is organised, central government remains responsible for allocating
transport funding across the regions. In doing so it is necessary
to distinguish between the overall benefits of transport spending
to the country and the economy as a whole, and the benefits to
individual regions and the objective of reducing regional imbalances.
36. If the objective is to maximise benefit to
the national economy, funds should be directed to where the potential
returns are highest - for example where congestion and overcrowding
are greatest. That may in practice mean disproportionate spending
in regions which are already richer and growing fastest - for
example London - and less in slower growth regions where pressures
on existing infrastructure are less. This applies whether the
DfT is allocating funds to lower tiers of government or prioritising
improvements to strategic national networks.
37. But insofar as the aim is to reduce regional
inequalities and give greater help to less buoyant regions, transport
funding can be allocated to them specifically to help stimulate
growth. But it should be recognised that this would probably mean
less benefit to the national economy as a whole. And the benefits
to the regions concerned may be weak unless the underlying conditions
for growth are favourable.
38. It is worth noting in this context that,
in the current appraisal methodology, use of common national values
of time in less buoyant regions, where prices and incomes are
typically lower than average, implies a bias in favour of these
regions. This is difficult to justify on efficiency grounds and
the overall benefits to economic growth, and has the effect of
giving additional weight to the regional equity objective.
Chris Riley
Chief Economist of DfT until 2005, now
working as a consultant
November 2010
113 "Transport, Wider
Economic Benefits and Impacts on GDP", DfT, 2005
Back
114
See the Appendix to Phil Goodwin's submission of evidence to this
Inquiry, September 2010. Back
115
"Feasibility Study of Road Pricing in the UK", DfT,
2004 Back
116
"Subsidising buses:how to get the best from taxpayers' money",
Oxera, June 2009 Back
117
"To Timbuktu and back again: why transport connectivity is
important", Oxera, October 2010 Back
118
See, for example: "Getting Transport Right", Campaign
for Better Transport and Green Alliance, 2008 Back
| 
