Docstoc

Energy Intensity

Document Sample
Energy Intensity Powered By Docstoc
					Energy Intensity: Complexities,
 Implications, Controversies

               Chris Green
         Department of Economics
            McGill University

 For presentation: Energy Institute, University of
      Wisconsin-Madison, March 10, 2011
                                                     1
     What is Energy Intensity?
• Energy intensity is the amount of energy used
  per unit of output.
• At the aggregate level it is E/GDP
• The components of EI change are:
  - energy efficiency (EE) change (or change in
  output per unit energy)
  - change in sectoral shares of energy and/or
  output, as EI varies across energy-using sectors
 Why the Interest in Energy Intensity (EI)?

• It is a key element of the Kaya Identity
  (next slide)

• The amount of carbon-free energy
  needed to stabilize climate is sensitive
  to the long-term (100 yr +) rate of EI
  decline (slide after next)
 Energy Intensity in the Kaya Identity
(1) C=P x GDP/P x E/GDP x C/E , where
 C= carbon emissions; GDP = world
 economic output; P=pop; E=energy.

We can convert eq (1) to rates of change by
 taking time derivatives of the natural logs
(2) %ΔC = %ΔP + %ΔGDP + %Δ(E/GDP) +
  %Δ(C/E)
                                               4
                      Energy Efficiency-Carbon-Free
                    Power Tradeoff (Hoffert et al, 1998)
                    60
                                     2100
                    50
Carbon-free power
  required (TW)



                    40      2050

                    30

                    20

                    10

                     0
                     0.0%    0.5%   1.0%    1.5%   2.0%   2.5%
                            Energy intensity decline             5
What Preceding Slide Indicates
If the:
 - 110 yr global GDP growth rate is ~2.4%,
  - Then the carbon-free energy required for
  atmospheric CO2 stabilization at 550 ppm is:

    ~ 37 TW for an EI rate of decline of 1.0%/yr
    ~ 19 TW for a rate of EI decline of 1.5%/yr
    ~ 7 TW for a rate of EI decline of 2.0%/yr
Energy Intensity is Controversial
•   Some believe that EE increase (and EI decline) are
    panaceas for stabilizing climate
     - Some emission scenarios (including many IPCC
    SRES scenarios) have very high rates of EI decline

     - But we shall see that even with very big energy
    efficiency increases, global EI decline is limited

•   Other believe “rebound” substantially reduces the
    effects of EE improvements on EI decline and on
    energy consumption (EC)
                                                         7
           How We Proceed
• Calculating Energy Intensity Decline

• Overstatement/understatements of the
  contribution of EE increase to EI decline
   - IPCC emission scenarios
   - So-called “rebound effect”

• Summary and Conclusion

Appendixes: A: EI in emerging economies: China
            B: Supporting Material
Calculating Global Energy Intensity Change
              is Complicated


• Many studies of EI decompose aggregate
  E/GDP into sectoral components
• But calculating overall EI from sectoral
  energy efficiency, GDP, and energy share
  estimates requires aggregation
• It turns out that the formula for doing so is
  complex with non-intuitive outcomes
                                                  9
    Implications of Complexity
• One outcome is that claims about potential EI
  decline are easily overstated (Baksi-Green,
  2007)


• As a result, these claims lead to overstatements
  of how much EE improvement can contribute to
  stabilizing climate



                                                  10
  Energy Intensity (EI) Calculations
• The next slide (from Baksi-Green (2007)
  presents:
  - sectoral estimates of attainable increases
  in energy efficiency
  - baseline (1990) estimates for energy (E)
  and GDP (Y) shares
  - ranges of future (2100) E and Y shares

                                             11
EI: Some Facts and Estimates




                               12
   Century-long Rates of Energy Intensity
                 Decline
• If we use the information in previous slide:
  -and assume plausible GDP shares in 2100
  - the 110-year rate of EI decline is 1.03%
• Now, if we double the rates of EE efficiency
  increase (except for electricity generation raised
  only to 85% and commercial raised to 600%),
   - the 110-year rate of EI decline is 1.25%

• Generally, the range for attainable EI decline is:
Range of Attainable Century-long EI
             Declines




                                      14
  Implications for Rate of EI Decline
• Achieving a 110-yr average annual rate of EI
  decline of over 1.0% will be difficult

• The required EE improvements to achieve a
  1.5% rate of EI decline over 110-yr are very
  (implausibly) great

• To achieve a 2.0% rate of EI decline over 110
  years requires astronomical rates of EE
  increase, ranging from ~500% to ~ 1100%

                                                 15
                And yet…


• Half of the IPCC emission scenarios build-
  into their baselines 110-year average
  annual rates of EI decline > 1.25%

• And three quarters have rates of 1.1% or
  higher.
                  Summary
•   Calculating the contribution of EE increase to
    EI decline involves aggregation

• What is true at the sectoral and country levels is
  not necessarily true of the “whole”

• We should not extrapolate from specific
  industries, sectors and polities to the world
 Implications for Climate Stabilization

• The next several slides indicate:
   - the size of the stabilization challenge
   - the existence of two “gaps”
   - the EI decline and carbon-emission free
  energy built into IPCC emission scenarios
   - the technology challenge of climate
  stabilization is huge

                                           18
  Emission Scenarios and the Technology
               Challenge
• We illustrate with the IPCC B2 scenario

• We distinguish two gaps;
   - one between the emission scenario and
  a stabilization path (in green)

  - one between the emission scenario and
 a ”frozen technology” baseline (in blue)
                                            19
        IPCC                 B2 scenario
                         Working Group III (2007)
        300
                                frozen technology
        250                           1990

        200

        150
GtCO2




        100
                                                    reference
         50                                          scenario

                                                    target level
         0
         1990   2010   2030 2050 2070 2090                20
               Various IPCC scenarios
                       Pielke et al, Nature (2008)
        7000


        6000


        5000


        4000
GtCO2




        3000


        2000


        1000


          0
               A1B   A1F1   A1T   A2     B1     B2   avg   med
                                               B2            21
         Breakdown of IPCC B2 scenario
        300
                               frozen technology
        250                          1990
                                                    sectoral
                                                      shift
        200                                         energy
                                                   intensity
                                      energy        decline
        150
GtCO2




                                     efficiency
                                   improvement
        100                                        carbon
                                                    free
                                                   energy
         50                                        current
                                                    policy
         0                                          focus
         1990   2010   2030 2050 2070 2090             22
         Breakdown of IPCC B2 scenario
        300
                               frozen technology
        250                          1990

        200
                                                   ~43 TW
                                      energy
        150
GtCO2




                                     efficiency
                                   improvement
        100                                         carbon
                                                     free
                                       ~26 TW       energy
         50                                         current
                                                     policy
         0                                           focus
         1990   2010   2030 2050 2070 2090              23
    Recapitulation to this Point
• The technological challenge to climate
  stabilization is huge
• Energy efficiency (EE) increase/EI decline
  is important but by no means sufficient
• There are physical/technical limits to EE
  increase, and these limit global EI decline
• Long-term rate of EI decline is sensitive to
  global share of energy intensive industry
                                             24
 But that is not the end of the story
• Just as some have claimed too much for EE
  improvement, others are claiming it won’t do
  much good at all
• They claim EE creates a “rebound” effect by
  raising the demand for energy. Why?
    - EE reduces cost (implicit real price) of
 energy, thereby raising quantity demanded
    - EE can contribute to economic growth,
 creating a (“derived”) demand for energy

                                                 25
   Rebound Effect and All That
• All true, but

  - other factors also contribute to
  growth, raising demand for energy.

  - not all of derived demand for energy
  should be counted as “rebound” from
  EE gains
    Energy and Economic Growth
•   Consider: economic growth is attributable to:

1. increased quantity of factors of production such
   as labor, capital and energy
2. increases in efficiency of these various factors,
   including those from economies of scale
3. other growth-inducing technological changes

•   EE may only be small part of growth equation
                                                    27
             Cause v. Effect
• Does EE cause growth or does growth increase
  the demand for energy?

• Some of both but, except where production is
  highly energy-intensive (as in Jevon’s The Coal
  Question, 1865), causation is likely to run from
  growth to energy

• Yet some “reboundists” attribute virtually all
  energy demand growth directly or indirectly to
  energy efficiency---terming it energy “backfire”
  What Does the Data Tell Us?
• Globally, GDP has grown faster than
  energy, and EI has declined as a result

• there is substantial variability over time in
  the relationship (see next slide)

• Growth can occur with and w/o EE
  improvement
                                                  29
 Average Annual Rates of Change
    in Energy GDP and EI (%)
Time period       Ê    Ŷ        Ê –Ŷ
 1900-2001a      2.4  2.9        -0.5
 1900-1950a      1.9  2.0        -0.1
 1950-2001a      3.1  3.8        -0.7
 1980-2000b      1.7  3.9        -2.2
 1994-2006b      2.3  3.3        -1.0
Sources: aAngus Maddison (2005); bEIA
^ = rate of growth of
 Assessing “Rebound” Requires a Baseline

• For any given GDP, the baseline should be
  what EC would be w/o EE increase

  (a) To begin: given global GDP, EE reduces
  energy consumption (EC)
  (b) The EC reduction is moderated by rebound
  (c) As the economy (GDP) grows EC rises
  (d) Typically, only a small part of growth-induced
  EC increase is due to EE–induced rebound
Rebound: The Baseline Matters
    What the Rebound Figure Tells Us
1. For given GDP, EE reduces EC by AC.
2. If GDP increases w/o EE increase, EC moves
   along AD
3. If there is GDP growth and there is an EE
   increase, then EC moves along CB

•    In either (2) or (3) an increase in GDP
     generates a derived increase in energy
     demand/EC

•    NB. DB indicates that with growth EC is lower
     than in absence of EE
Making CF the Baseline Creates a Straw Man
  Straw Men and Non-Sequiturs
• We can now see that the rebound debate is
  based on a “straw man” and a non-sequitur
   - Straw man: assumes that in absence of
  rebound EE improvements imply EC decreases
   - Non-sequitur: attaches most (if not all) of the
  aggregate output-induced increases in EC
  either directly or indirectly to EE improvement
• In fact, regardless of rebound, over time EE
  reduces EI, and keeps the growth in EC lower
  than it would otherwise be
    Summary and Conclusions
• EI decline is ubiquitous and important

• But there are limits to EE increase/EI decline
  implying that:

1. EI decline is a necessary but not nearly
   sufficient condition for climate stabilization
2. The insufficiency is mainly physical/technical not
   behavioral (“rebound”)
                                                    36
 The Lesson of Energy Intensity
• The real lesson of EI decline is that without a
  technology revolution in energy supply climate
  cannot be stabilized

• The reason is not EE rebound; rather a growing
  world economy needs more energy

• Energy efficiency gains can slow that growth but
  not reverse it.

                                                    37
  Appendix A: The China Story
• Climate change is not the only great event
  of the 21st century
• Economic transformation of once poor
  countries is another
• Much of that development is energy
  intensive---some of it necessarily so, a
  factor that will limit global EI decline
• China provides an example
                                           38
the China model




             39
         Urbanization and EI
• Urbanization is an important facet of
  China’s economic transformation
• In populous countries this implies high-rise
  buildings and accompanying infrastructure
• These require energy-intensive materials
• And without a revolution in energy
  technologies rapidly rising emissions
• the next three slides illustrate
                                             40
China’s Share of
                  global production              (2006)

     Flat Glass                            49%

       Cement                              48%

          Steel                      35%

    Aluminum                   28%

 Passenger cars          11%

      Ethylene      8%

Semiconductors     7%

                                                      41
          EI Change in China
• It is also instructive to look at China
• China’s growth has been accompanied by very
  different rates of EI decline over last 30 years
• This can be inferred from the carbon intensity
  (C/GDP) change line (next slide) since:
     - most of the decline in C/GDP was due to
       EI decline, and
    - most of the EI decline was due to EE
       decline.
China’s Carbon Intensity
  (Zeng, et al, Science, Feb 8 2008)
      Per capita CO2 emissions
                    2004 vs 2030
25




                            .6
                            .7
                         19
                         19
20

15




                                               9
                                              5
                                            9.
                                            9.
10
     6
          2




                    1
     7.
          7.




                    7.
               7




 5
               3.




                                        8
                                     1.
0                                1
      EU       China      US     India      Japan   44
        Emission Projections
12000


10000                                          China


8000


6000                     US


4000                    EU


                                                India
2000
                   Japan

   0
   1974   1982   1990    1998   2006   2014   2022   2030 45
  Appendix B: Supporting Material
• The next two slides calculate 110 year (1990-
  2100) rates of EI decline
   -the two slides differ in the sectoral energy
  efficiency increases
  -while using the same future GDP shares
• Doubling the increase in energy efficiency in all
  sectors except electricity generation only raises
  the EI average annual rate of decline from
  1.03% to 1.25%

                                                      46
Energy Intensity Decline Rate
    1990-2100 = 1.03%




                                47
A 1.25% rate of E/GDP Decline




                            48
         The Next Two Slides
• The first slide indicates the sensitivity of 110
  yr rates of EI decline to future (2100) E and
  GDP shares
   - there is a wide range of results

   - but the range is much smaller when only
  plausible future E and Y shares used

• The second slide shows the rates of EI
  decline implied by IPCC emission scenarios
                                                 49
Sensitivity Analysis:
 A,B,C likely; D,E,F Unlikely




                                50
Calculated Rates of EI Decline
    Based on IPCC SRES




                                 51
        IPCC                  B2 scenario
                          Working Group III (2007)
        300
                                frozen technology
        250                           1990

        200
GtCO2




        150

        100
                                                     reference
         50                                           scenario

                                                     target level
         0
         1990   2010   2030   2050   2070   2090           52
Contributions to CO2 Emissions
       IPCC 2007 WGIII TS p.6
Advanced Energy Technology Gap
• Given the clean energy gap, we can ask
  how much of that gap requires energy
  technology breakthroughs
• A qualitative analysis is found in Hoffert et
  al, Science (2002).
• Green, et al (2007) attempt a quantitative
  analysis, as indicated in next slide.


                                              54
55
 EI and Global Trade in Carbon
• Export-oriented growth accounts for 20%+
  of China’s total carbon emissions
• China’s (EU’s) carbon “footprint” is over-
  (under)-stated on a consumption basis
• Kyoto-based emission production targets
  may have increased global emissions
• Another example of what may be true for
  individual is not true of whole
                                           56
  Carbon Dioxide Transfers
(Davis and Caldeira, PNAS, March 2010)




                                         57
 Prospects and Policies for EI Decline

• The prospects for continued EI decline are good
 - improvements in energy efficiency (EE) are
  ubiquitous and market forces put a premium on
  their eventual adoption
 - scientific breakthroughs will also be required
 - but ultimately there are limits to EE increase
 - these limits, plus future E and Y shares of
  energy intensive sectors, affect whether a 1.0%,
  century-long rate of decline in EI is achievable


                                                 58
 Policies to Promote EE Improvement

• The favored approach among economists is a
  price on carbon
• This would both induce conservation and the
  use of energy intensive alternatives
• A high price of carbon is politically toxic, and
  unlikely to induce technological breakthroughs
• In some cases (appliances new buildings)
  efficiency standards are more effective
• A combination of standards, low carbon tax, and
  a technology policy is a better approach

                                                 59

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:8
posted:3/28/2011
language:English
pages:59