Paul Watkiss

1
Carbon Policy Appraisal in the UK
 
Paul Watkiss paul_watkiss_at_btinternet.com June
2008
2
Presentation Outline

• The development of the UK policy and estimates
• SCC to SPC .to MAC ?
• The social cost of carbon some issues
• Applications
• Types of application (policy vs. project
  appraisal)
• Example use of a value of the UK values

3
Definitions

• Social cost of climate change economic costs of
  climate change effects, usually for the business
  as usual (no mitigation). Also known as cost of
  inaction
• The marginal social cost of climate change
  (social cost of carbon, SCC) is the marginal
  damage from 1 additional tonne of emissions,
  either as CO2 or C
• It is the net present value of climate change
  impacts over the next 100 years (or longer) of
  one additional tonne of carbon emitted to the
  atmosphere today
• It can be compared to the marginal abatement
  costs (MAC) of mitigation to look at climate
  change policy, or used as a shadow price in
  project / policy appraisal
• Both SCC and MAC should not be confused with the
  traded price of carbon, which is MAC for
  short-term pre-defined emission caps

4
The UK SCC Value

• In 2002, the UK Government Economic Service
  recommended a SCC
• Based on review of literature of marginal SCC
  values (ExernE - Tol / Downing) 1
• Illustrative central 70/tonne of carbon (tC),
  range 35 to 140/tC, for year 2000
• (10 19 - 38/tCO2) or (Euro 12 25 - 50/tCO2
  ER 1.3)
• These values much higher than other
  countries/organisations at the time
• Rising at 1/tC/yr in appraisal
• Recommended to value carbon in project and
  policy appraisal in Government.

5
Government Applications SCC
GHG Mitigation/ Climate change policy
Other Policy, e.g. Transport, Energy
Future Transport appraisal (DfT) Energy
Investment (Ofgem) APM4 (EA)
Project appraisal
Energy White Paper (DTI) Renewables Obligation II
(DTI) NTM/SP model (DfT)
F Gas RIA (Defra) CBA UK ETS
Policy appraisal (RIA)
Aviation Tax Consultation (Dft) RUC/DD analysis
(DfT) Waste taxes/charges (Defra) Building
regulations (ODPM)
Economic Instruments
Climate change evaluation Revised CC programmme
Long-term CC Goals
Can involve increase or decrease in GHG
6
The UK SCC Review

• Government commissioned a review of SCC in
  2004-2005
• One reviewing values (Downing, Tol, Hepburn,
  Hope, Watkiss, Anthoff, et al )
• One reviewing use of values in policy (Watkiss,
  et al)
• Different lines of evidence followed including
  review, new modelling work (2 models FUND and
  PAGE), analysis, consultation

7
The UK SCC Review Findings

• Understanding is improving, but still poor
• Estimates of the social cost of carbon span at
  least three orders of magnitude, from zero (or
  even below) to over 1000/tC
• Reflects uncertainties in climate and impacts,
  coverage of sectors and extremes, and choices of
  decision variables (notably discount rate, equity
  weighting, risk aversion).
• Lower UK SCC benchmark of 35/tC is reasonable,
  but
• for global decision context committed to reducing
  the threat of dangerous climate change and
  including modest level of aversion to extreme
  risks, relatively low discount rates and equity
  weighting.
• Upper benchmark is more difficult to deduce
  (though risk of high values is significant).
• Not possible to put forward a central value with
  any confidence

8
The UK SCC Review - Policy

• Economic benefits of climate change should be
  considered in long-term targets, however, a wider
  framework is needed to take all relevant effects
  into account
• Detailed policies follow from, and should be
  consistent with the long-term goals.
• Day-to-day policy decisions (policy or project
  appraisal, or setting levels of taxes and
  charges), should reflect the policy commitments.
  
• Should ensure that the (long-term) target is
  achieved in a cost-effective way, so need
  consistency across all applications in
  Government.
• Two-tier approach with different analysis for
  long-term targets (multi-attribute framework)
  short-term economic appraisal (shadow prices for
  day-to-day appraisal).
• no single SCC estimate, method, model or tool
  adequately captures all of the uncertainties).
  Single estimates of the SCC should be avoided for
  (long-term) policy decisions. Disaggregated
  analysis of economic winners and losers by region
  and sector, and impacts of climate change
  including key indicators such as health and
  ecosystems. Consideration of all risks, including
  major events, discontinuities and large socially
  contingent events (risk matrix). Extensive
  sensitivity analysis.

9
The UK SCC Review - Lessons

• Stakeholder consultation reviewed wide views.
  Widespread disagreement on how best to set long
  term climate policy, but most agreed need for
  shadow prices in day to day appraisal.
• Disagreement was on how to set these. 1) MAC, 2)
  SCC, or 3) pragmatic approach.
• Determined by existing perception about welfare
  economics, climate change (is it a special case?)
  and confidence in estimates of the SCC
• Issue of uncertainty. Note that review of MAC
  shows at least an order of magnitude difference
  in estimates of MAC for given emission reductions
  targets
• Review used a pragamatic approach to produce an
  illustrative set of values. Use of HMT declining
  discount rates, equity weighting, mean values.
  Combined with consideration of MACs. Broadly
  similar to original SCC values
• For day to day appraisal, conflict between
  reflecting uncertainty (range of values) versus
  consistency (single value)
• In some respects, it does not really matter what
  the value for appraisal is, as long as there is a
  value, and it is applied consistently.

10
Stern Review

• Used one model (PAGE) to estimate social cost of
  carbon
• Ran with pre-defined assumptions on discount rate
  (0.1 prtp), factored in an equity adjustment
• Business as usual SCC value 85/tCO2
• Also looked at value 450 550 ppm CO2e
  stabilisation, i.e. SCC on different trajectories

11
UK Gov Shadow Price of Carbon (SPC)

• 2007. Defra reviewed evidence. For consistency
  Stern value, but used 550 ppm CO2e stabilisation
  value 30/tCO2 (2000)
• Converted to 19/tCO2e. Uprated to 2007 value
  2007 emission 25/tCO2e.
• Decided to work with a central value, not a range
• Value increases at 2 a year for use in appraisal
• Then discounted using standard HMT 3.5 discount
  rate, or if gt30 years, the declining discount
  rate scheme
• In paper, however, it says the SPC is different
  from the SCC as can adjust to reflect the policy
  and technological environment
• And can reflect that may affect UK willingness
  to pay for reductions in carbon emissions, such
  as political desire to show leadership in
  tackling climate change.

12
Climate Change Bill and Committee (CCC)

• Interim carbon budgets for the UK, five year
  periods, 2008 2022
• CCC commissioned review on how to set these
  targets (Watkiss, Dietz, Blyth)
• Recommendation was that interim targets need to
  be consistent with long-term target, but include
  option value to allow future changes to long-term
• SPC too uncertain to use in long-term and interim
  target setting for optimisation
• Build analysis around UK specific MAC, as part
  wider multi-attribute framework
• Raised issue of consistency with this UK MAC
  based approach vs. the use of a SPC derived from
  SCC for appraisal in government
• And of course the issue of consistency and
  influence EU targets
• Note that UK MAC determined by how much
  domestically, and also traded price
• Defra possible move to MAC for SPC, probably UK
  MAC consistent with CCC

13
Observations

• The SCC value in the UK has remained pretty
  constant over past 6 years, 3 reviews, despite
  large changes in the literature estimates (down
  and up). Reduces confidence if numbers change
  regularly.
• The SCC, for policy, seems to have a goldilocks
  quality. It should be neither too low or too
  high.
• No evidence of reverse engineering values, i.e.
  picking the number wanted, and then justifying
  the parameter set, trajectory (dr) but
• The justifications on parameter choice have
  changed over time.
• Choices are different to other UK appraisal, 1)
  global not UK 2) Equity-concerned global policy
  maker. Not consistent with other policy (trade
  aid).
• The discussion following Stern shows that there
  is no consensus on parameter choice, let alone
  values

14
Parameter choice

• Recent runs (MethodEx) with FUND
• If one adopts a conventional marginal economic
  appraisal perspective (desc.)
• Standard discount rates (4 SRTP in Europe 2
  PRTP)
• No equity weighting
• Standard approach to risk median values (or
  best guess)
• Gives a low value

15
Parameter Choice (2)

• If one adopts a different decision perspective
  (e.g. concerned global policy maker),
  prescriptive
• Low discount rates (declining or 0-1 PTRP )
• Consideration of equity weights
• Greater consideration of risk mean values
• Higher values - but still low in relation to
  long term MACs

16
Discounting
Source FUND
17
Equity Distributional Effects
IMPACTS VS ADAPTIVE CAPACITY
MOST VULNERABLE
RESIDUAL RISKS

• Poorer countries likely to be net losers, as more
  vulnerable
• climate-sensitive activities, close to
  tolerances, poorly prepared to adapt

How adjust impacts in regions (equity weighting)
18
The Effect of Equity Weighting
Source FUND
Note PAGE shows lower values if equity weight as
dynamic weights over time Consistency with other
policy
19
Coverage
Socially
Market

Non
-
Market

contingent
Limit of coverage of many studies
Projection e.g. temperature or SLR
None

Limits of coverage for most studies
Bounded e.g. precipitation and extremes
None
None
Major change
One or two studies

None
None
e.g. major tipping points
Source Watkiss et al, 2006
Missing elements will include positives and
negatives, but more likely to be negative
20
Applications for an SCC

• Long-term (stabilisation) policies
• National Interim target setting (e.g. carbon
  budgets 2020)
• Economic instruments
• Policy appraisal
• Project appraisal
• Note for policy and project appraisal, likely to
  be a combination of policies and project that
  increase as well as decrease greenhouse gas
  emissions
• Adaptation
• though greater consideration of costs of inaction
  at national level, for different scenarios (not a
  global SCC value)

21
Current project and policy appraisal

• http//www.defra.gov.uk/environment/climatechange/
  research/carboncost/index.htm
• 19/tCO2 in 2000 prices for 2000 emission (from
  Stern 550).
• 1) build in observed inflation to get to 2007
  value, 25.5/tCO2 in 2007 prices
• 2) Uprated 2 per year future years (rising
  damage costs from higher greenhouse gas
  concentrations)
• Then discount using standard 3.5 discount rate,
  or declining discount rate scheme if gt 30 year
  scheme
• Some additional guidance prudent to look at
  sensitivity test with /-5, though also some
  reference in some RIAs to -10 to 20

22
Current Example

• Heathrow third runway (Adding capacity) considers
  three options. http//www.dft.gov.uk/consultations
  /closed/heathrowconsultation/consultationdocument/
  annxb.pdf
• E.g. option 1 involve significant capital and
  refurbishment costs associated with new runway,
  terminal and surface access infrastructure
  estimated at 6.8bn 7.6bn.
• Additional non-infrastructure costs in terms of
  climate change 4.8bn (and air noise costs are
  estimated at 0.3bn).
• There are significant transport user benefits
  from additional capacity estimated at 17.1bn.
  NET BENEFIT (NPV Best estimate) 4.8bn
• Approach is a little complicated, as looks at
  emissions over time, and applies an aircraft
  uplift factor for non-ground CO2 (1.9). Options 2
  and 3 take account of benefits from reduced
  delays. Discounted using declining discount rate
  scheme

23
(No Transcript)
24
70 years
25
(No Transcript)
26
But

• Some external reviewers have commentated that
  globally we are not on a 550 ppm trajectory.
  Therefore, the SPC should be higher
• If work with something closer to the Stern BAU
  (85/tCO2), increases CC cost of Heathrow option
  1 by so much that it would remove the positive
  NPV for the project
• This highlights the issues of ranges of values
  vs. consistency
• The sensitivity test recommended /- 5 (for HRW
  -10/20) is not reflective of the range of
  values, merely an arbitrary sensitivity analysis

27
Additional Issues

• Other greenhouse gases. Guidance recommends the
  use of GWP.
• Strictly speaking this is incorrect, as value of
  gases do not scale in this way because of
  atmospheric lifetime and discounting, but
  assumption on what should use depends with other
  assumptions (so complex).
• The rate of uplift. Recommends 2 real.
• Hope argues probably closer to 3, but in reality
  varies by sector, and also varies dynamically
  over time (non-linear). However, 2 is
  consistent with other areas of policy (e.g.
  health 2 uprated)
• Trajectory. Assumes 550 ppm trajectory
• Stern recommends 450-550. 550 is conservative,
  in that produces a higher value than 450
  (probably).

28
Observations

• Our understanding of the SCC is still evolving.
  There are very large uncertainties from decision
  parameters and choices so range of values at
  least two orders of magnitude
• The choice is determined by perspective.
• With a standard decision approach, low numbers.
• If consider climate change special, and adopt a
  different perspective, then can derive higher or
  high numbers. However, then applying different
  approach to other policy.
• There is no consensus, only individual viewpoints
  !
• Our understanding of MAC is still evolving. Note
  that switching to a MAC does not remove
  uncertainty. Very wide range of values on MAC
  (e.g. Barker et al) that vary even with sign
• At same time, emergence of carbon prices
  determined by political negotiations interested
  in a shadow price for cost-effective achievement
  of these targets

29
The Policy Problem

• Long-term political targets, set on precautionary
  principle, not economic assessment
• EUs 2 degrees target (in the UK, at least
  a 60 CO2 reduction by 2050)
• There are potential (multiple) pathways
  consistent with such targets (allocation and
  trajectory). Therefore, multiple variants on
  MACs at interim dates towards any target
• But also short-term political targets (2020,
  203), set on the basis of headline policy
  aspirations, not on economic assessment or CBA
• Sit within a plausible trajectory to long-term
  target, but so do many other variations
• 203 supported by MAC analysis that aim to
  demonstrate that the prices associated with
  achieving these targets are not
  dis-proportionately high (e.g. EC IA, 40
  Euro/tCO2)
• The problem is how to set short-term values for
  appraisal, within this policy framework of
  interim and long-term targets
• And whether goal is instead to achieve mandatory
  political targets cost-effectively
• Plus covering finance streams from possible Gov
  expenditure for flexible mech

30
Final thoughts

• Important to separate out the application of
  long-term policies from day to day policy
  appraisal. However, two are linked in
  considering SCC or MAC and policy targets
• Once set a number, seems to acquire status and is
  difficult to change, but do need to build in
  formal review process
• Practical issue of whether singe estimate or
  range. Merits in both, and choice is over
  consistency vs. uncertainty.
• Possible to select a SCC value, but unlikely to
  get consensus on SCC value / parameter choices.
  Issue of consistency with other policy areas on
  these choices.
• Many argue for MACs as an alternative, but not
  remove uncertainty
• Possible to use a pragmatic choice that merges
  the two, but should recognise that difficult to
  provide a theoretical justification for this.
• Inconsistency if trying to achieve a EU wide
  target that itself has been set through political
  process (203), which will need consideration of
  country specific MACs, traded prices, etc
• The best approach is likely to be determined by
  the overall ambition of implementing shadow
  prices a general price signal vs. ensuring
  (203) cost-effectively vs. other

31
(No Transcript)
32
Moving to climate policy Interim CCC

• Recent study for CCC
• The aim of the work has been to review the
  different methodological approaches for setting
  interim UK carbon budgets for OCC / SCCC

• Supports role of the Committee on Climate Change
  (CCC), to provide advice on the first three
  carbon budgets (2008-12, 2013-17, 2018- 22) by
  Sep. 2008.
• 2020 target is bounded by EWP 2003 26 32 CO2
  reduction on 1990, and at least a 60 by 2050
• Also informed by Stern Review (450 550 CO2e
  ppm) other commitments (EU 20/30 by 2020)

Roles of the CCC
33
Targets

• Possible (logical) flow from global to
  sub-sectoral
• But interim targets linked to top-down and
  bottom-up influences
• All decisions at all stages subject to discussion
• Flow upwards is determined by action of others
• Interim target has to be considered in terms of
  overall picture (not in isolation)

34
(No Transcript)