Transcript Natural resource consumption, technological progress and future

On Economic Growth, Energy
Consumption and Technological
Change
Jussieu 24 Avril 2006
Dr Benjamin Warr
Professor Robert Ayres
Introduction to INSEAD
• Two fully connected campuses in Asia
(Singapore) and Europe (France), 143
faculty members from 31 countries, 880
MBA participants, 55 executive MBAs,
over 7000 executives and 64 PhD
candidates. On both campuses, faculty
conduct leading edge research projects
with the support of 17 Centres of
Excellence.
Sommaire
• Critique de l’approche «neo-classique » de
la croissance économique
• Considération de la rôle d’énergie
• Estimation d’une « proxy » mesure de
Technologie
• Développement d’une méthode pour
estimer la croissance du Produit Intérieur
Brut.
Problématique
• L’approche neo-classique économique
– Ignore l’environnement et des ressources
naturelles
• Comme facteur de production
• Comme bien collectif
– Considère la technologie comme exogène,
continue et perpétuelle.
• Mais le progrès technologique est plutôt non
linéaire (learning by doing) avec des limites
Une fonction de production
• Décrit les relations entre le « output »
(PIB) et les « inputs », (les facteurs de
production)
• Cobb-Douglas ont développe la forme le
plus utilisé,
Y = A KL
where  +  = 1
• Y=PIB, A=technology multiplier, K=capital,
L=labour,  et  les élasticités de
production
Quelques problèmes
• Les ressources naturelles exclus….
• Constant returns to scale (rendement constant)
• Le dérivative défini la productivité marginal de
chaque facteur en tant que constant, égal au
« factor cost »  =0.3 capital,  =0.7 labour.
• Static substitution
• Rendu dynamique avec multiplicateur
technologie (A), l’erreur d’une modèle OLS.
• PAS de RETROACTION suites aux
changements dans le quantité et qualité du
bilan énergétique.
PIB et les facteurs de production, K, L, B, US 1900-2000
25
index (1900=1)
20
PIB (Y)
Capital (K)
Labour (L)
Ressources Naturelles (B)
15
10
5
0
1910
1920
1930
1940
1950
année
1960
1970
1980
1990
2000
PIB empirique et estimé, et l'erreur (le progres technologique)
25
PIB empirique
20
index (1900=1)
PIB estimé (Cobb-Douglas)
Erreur (technological progress)
15
10
5
0
1910
1920
1930
1940
1950
année
1960
1970
1980
1990
2000
Observations
• Même avec inclusion des ressources naturelles
(B) le PIB estimé est inférieur au valeur
empirique si on utilise les « factor costs » pour
définir les paramètres.
• Le progrès technologique (l’erreur) est
responsable pour plus que 80% de la
croissance.
• Si on utilise pour prévision on est obligé de faire
l’hypothèse que la technologie va développer
comme avant. La croissance économique est
assuré malgré nos actions.
Industrial Metabolism
(Ayres and Simonis 1994)
• New conceptualisation of society’s relation to and
pressures on the environment.
• The economy is physically embedded into the
environment.
• The economy is an open-system with regards matter &
energy.
• Matter and energy societal throughputs must =>
minimum requirements = technological progress.
• RESOURCE SCARCITY: Societies intervene with
purpose to gain better access to supplies of natural
resources (through technology and resource
substitutions .i.e. energy) – a supply-side problem.
• ASSIMILATIVE CAPACITY: Societies must restrict
waste flows to the environment (output side).
The Salter Cycle, an engine for growth.
Product
Improvement
R&D Substitution of
Knowledge for Labour;
Capital; and Exergy
Process
Improvement
INCREASED REVENUES
Increased Demand for
Final Goods and Services
Substitution of
Exergy for Labour
Lower Limits to
and Capital
Costs of
Production
Economies of
Scale
Lower Prices of
Materials &
Energy
Criteria for Environmental
Accounting
• Environmental accounting must be:
– Politically relevant – strength of the concept to
provide information for policy decision and
public discourse.
– Feasibility often requires reduced complexity
– Definition of scale and then system
boundaries
– Accurate source information
– Methods to estimate stocks & flows
Energie comme facteur de
production – quel mesure faut il?
• Pas tout l’énergie utilisé est utile dans
l’économie – conséquence du 2eme loi de
Thermodynamique.
• Faut considérer la quantité plus qualité de
l’énergie utilisé
• Faut quantifier le progrès technologique et
l’effet sur la quantité et le façon qu’on
utilise énergie.
Task efficiency: specify service &
define the task
• The first objective of any technical study of energy use is
to establish a standard of performance.
• What is the difference between a service and a task?
– (service) keeping warm, (task) providing heat to a home
– (service) structures in society, (task) making aluminium
– (service) mobility, (task) moving a vehicle
• Services must consider non-technical trade-offs, tasks
require only a physics perspective.
• This permits,
a) Evaluation of the efficiency of present uses.
b) Definition of goals towards which technical
innovation can strive.
Thermodynamics and
« available work »
Necessary to define a Minimum Task Energy to
allow consideration of :
• Interchanging devices or systems (mass transport vs. Cars)
• Seeking technological innovations (aluminium for steel)
• The 1st Law (convervation of energy) is
inadequate for considering minimimum task
energy.
• The 2nd law (the entropy law) indicates that « in
any process involving heat, there is an
inexorable increase of entropy (disorder),
meaning that not all the energy is available in
useful form »
The 1st Law (conservation of energy) is
inadequate for considering minimimum
task energy.
• η = energy transfer (of desired kind) /
energy input
• Maximum value may be greater than 1.
• No explicit consideration of the quality of
the energy and its ability to do useful work.
• Cannot be generalised to complex
systems with work and heat outputs.
The 2nd law (the entropy law)
• indicates that « in any process involving heat,
there is an inexorable increase of entropy
(disorder), meaning that not all the energy is
available in useful form »
• For any device or system the 2nd Law Efficiency
ε is the ratio of the minimum exergy that could
perform the task (Bmin), to the exergy actually
consumed in doing the job (Bactual).
• Its maximum value is 1.
• Maximising ε minimises exergy demand and
wastes generated for a given task.
Exergy and Exergy Balance
• Exergy is the useful part of the energy.
• There are 4 components:
– Kinetic exergy of bulk motion
– Potential gravitational or electro-magnetic field
differentials
– Physical exergy from temperature and pressure
differentials
– Chemical exergy arising from differences in
chemical composition
• We can ignore the first two for many industrial
and economic applications.
Exergy or « Available Work »
• So, not all energy can be made available in useful form
(consequence of 2nd Law).
• Available work is an energy measure that is actually
consumed in a process.
• Work is the highest quality (lowest entropy) form of
energy. It is often called exergy.
• Exergy = The maximum amount of work that a
subsystem can do on it’s surroundings as it approaches
thermodynamic equilibrium reversibly.
• Exergy is proportional to the future entropy production,
but has units of energy.
• Exergy is gained or lost in physical processes.
• Minimising exergy consumption is a measureable
objective to optimise energy consuming tasks.
Example: Chemical exergy
•
•
•
•
Production of pure iron (Fe2) from iron oxide
(Fe2O3)
This requires exergy from burning coke (pure
carbon)
Carbon dioxide (CO2) is the waste product
2Fe2O3 + 3C  4Fe + 3CO2
Correct mass balance – all atoms in ome out.
Conversion of mass causes inevitable joint
product CO2
0.75 moles of CO2 per Kg of Fe.
Iron production 1
1.
2.
3.
4.
5.
2Fe2O3 + 3C  4Fe +
3CO2
Making 4 moles of Fe
requires generation of 3
moles of CO2
And 1505.6 Kj which
comes from this
oxidation of carbon
But 3 moles of C
contain only 1230.9
We need 0.76 C extra.
Weight
kJ/mole
exergy
Fe
56
376.4
Fe2O3
160
16.5
C
12
410.3
CO2
44
19.9
O2
32
4.4
Iron Production 2
2Fe2O3 + 3C  4Fe + 3CO2
Correct mass balance, incorrect exergy balance
2 Fe2O3 + 3.76 C + 0.76 O2  4 Fe + 3.76 CO2
(33.0)
(1542.7)
(3.0)
(1505.6) (74.8)
On the input side oxygen has been added to fulfill the balance of the
extra C required
1580 kJ in  1580 kJ out
• This is for an ideal reversible transformation. No entropy
generated or exergy lost.
• Hence 0.94 moles of waste CO2 are inevitable per mole
Fe produced (corresponds to 0.74kg CO2 per kg Fe)
• This is the thermodynamic minimum.
Iron Production: Reality
• The 410.3 kJ/mole from source C is never used
100% efficiently
• Blast furnace average have efficiencies of 33%.
• So, one mole of C one obtains only 135.4kJ
• As a result need 12.42 moles of C instead of
3.76.
2 Fe2O3 + 12.42 C + 9.42 O2  4 Fe + 12.42 CO2+ heat
(33.0) (5095.9) (37.7) (1505.6) (247.2)
• B lost = 3413.8 kJ
• 2/3 rd of waste produced is unecessary.
Types of Exergy Service
•
•
•
•
•
•
Prime Movers ( electricity)
Transport
High Temperature Process Heat
Mid and Low Temperature Process Heat
Lighting
Non-Fuel
Petroleum Products
Apparent Consumption
Gasoline
Petroleum Exergy
Flows
Diesel
Transport
Aviation Fuel
Electricity
Furnace Oil
Heavy Fuel Oil
Space Heating
Process Heat
Transport
Kerosene
Lighting
Petroleum Coke
Process Heat
Feedstock
Non Fuel
Bitumen/ Waxes
LPG
Allocated to gas flows
Coal, Petroleum, Gas: Exergy breakdown by use, US
1900-2000
Transport uses
Figure 9. Petroleum and NGL consumption: Exergy
allocation among types of work, USA 1900-1998
Figure 8. Coal consumption: Exergy allocation among
types of work, USA 1900-1998
Figure 10. Natural Gas consumption: Exergy allocation
among types of work, USA 1900-1998
90%
100%
HEAT
ELECTRICITY
PRIME MOVERS
NON-FUEL
90%
80%
80%
100%
HEAT
ELECTRICITY
PRIME MOVERS
NON-FUEL
LIGHT
90%
80%
70%
70%
70%
60%
60%
50%
50%
Fraction (%)
Fraction (%)
60%
Fraction (%)
HEAT
ELECTRICITY
PRIME MOVERS
NON-FUEL
40%
50%
40%
40%
30%
30%
30%
20%
20%
20%
10%
10%
10%
0%
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
0%
0%
1910
1920
1930
year
Declining fraction
to heat
1940
1950
1960
1970
1980
1990
2000
1910
1920
1930
1940
year
Increasing fraction
to electricity
1950
year
1960
1970
1980
1990
2000
Figure 11: Percent of fossil fuel exergy consumed by
types of enduse, USA 1900-1998
Total Exergy Breakdown by Use, US 1900-2000
90%
HEAT
ELECTRICITY
PRIME MOVERS
NON-FUEL
LIGHT
80%
70%
Heat
Fraction (%)
60%
50%
40%
30%
Electricity
Other Prime
Movers
20%
10%
Non-Fuel
0%
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
2000
Estimated historical efficiency of lighting, USA 1900-2000
3.50%
Lighting Efficiency
3.00%
efficiency
2.50%
2.00%
1.50%
1.00%
0.50%
0.00%
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
Efficiency (%)
1900
1950
1972
2000
Kerosene
1%
20%
5%
1%
1%
year
Efficiency (%)
Incandescent Fluorescent
5%
15%
Market Share (%)
80%
0%
70%
25%
65%
33%
60%
39%
Average Efficiency
1.400%
2.433%
2.737%
2.953%
Bauxite Ore
3.9kg (4.1MJ)
Refining
Coal, Oil, Gas
1900 = 82 MJ/kg
2000 = 28 MJ/kg
Electrolysis
Electricity
1900 = 190 MJ/kg
2000 = 66 MJ/kg
Casting
Cokes
1900 = 20 MJ/kg
2000 = 10 MJ/kg
Aluminium
1kg (32.8MJ)
Coal
Oil
Gas
Electricity
Total
MJ/1000kg
4092
10912
8281
56559
79845
Simplified
process view:
Aluminium
% of total
5%
14%
11%
70%
100.00%
Table 1. Breakdown of total fuel exergy inputs for the production of 1 ton of primary
aluminium (source: IAI LCS 2000).
Exergy consumption per kg of Al produced
250
Bauxite
Coke
Coal, oil and gas
200
Electricity
MJ/kg
150
100
50
0
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
year
Figure 1. Exergy consumed per kg primary aluminium produced. *electricity consumption
adapted from Energy Implications of the Changing World of Aluminium Metal Supply (JOM 2004).
Efficiencies and GDP/Exergy Input
Figure 13. Energy (exergy) conversion efficiencies,
USA 1900-1998
40%
35%
Electric Power
Generation &
Distribution
30%
efficiency
25%
High Temperature
Industrial Heat
20%
15%
10%
5%
Medium Temperature
Industrial Heat
Mechanical Work
Low Temperature Space Heating
0%
1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
year
Technical
efficiency, US 1900-2000
Figure 3. Technical efficiency learning curve model,
USA 1900- 2000.
0 ,18
0 ,16
technica l efficiency, f
0 ,14
0 ,12
0,1
0 ,08
0 ,06
emp ir ica l (U/ R)"
0 ,04
b ilog istic mod el
0 ,02
0
25
6 95
14 86
26 60
4 6 77
cumula tive prima ry e xe rgy prod uction (eJ)
Source Data : Ayres, Ayres and Warr, 2003
7 11 3
Useful Work/GDP Ratios,
US 1900-2000
Figure 14. Primary work and the primary work / GDP
ratio,
USA 1900-1998
5.0
work (Ue) / GDP ratio
4.5
work (Ub) / GDP ratio
4.0
3.5
ratio
3.0
1st Oil
Crisis - US
Peak Oil
Production
2.5
2.0
1.5
1.0
0.5
0.0
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
2000
How does our model work ?
Kt t , Q
Yt  QY
Lt,K
Rt ,,LtCobb-Douglas
, Rt ,
or LINEX
Yt  KYtt LKt t  F
Lt Rt t  Ft RKt t Lt KUtt Lt U t




   L  U 
 L 
Yt  U exp a 2  
   ab  1
 U 
   K 
• At the ‘total factor productivity’ is REMOVED
• Rt natural resource services replaced by Useful
Work, where U = F * R
• Ft technical efficiency of energy to work conversion
REXS economic output module
Exe rgy
Serv ice s
Labour
Linex
parameter a
Gross Output
Capital
ICT Fraction of
Capital
Linex
parameter b
ICT Capital
Growth Rate
Linex
Parameter c
Monetary
Output
Cumulativ e
Production
Monetary
Labour supply feedback dynamics
Labour
Labour Fire
Rate
Labour Hire
Rate
<Time>
Fr actional
Labour Hire Rate
A
Structural Shift
Time C
Fr actional
Labour Hire Rate
B
Fr actional
Labour Fire Rate
A
Structural Shift
Time D
Fr actional
Labour Fire Rate
B
Parameters for USA 1900-2000
• Structural Shift Time C=1959, Structural Shift Time D=1920
• F Labour Fire Rate A=0.108, F Labour Fire Rate B=0.120
• F Labour Hire Rate A=0.124 F Labour Hire Rate B=0.135
Labour “hire and fire” parameters
Simulated labour hire and fire rate, USA 1900-2000
0,45
0,4
Labour Hire Rate
rate (standardised labour units per year)
Labour Fire Rate
0,35
0,3
0,25
0,2
0,15
0,1
0,05
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
2000
Labour – validation by empirical fit
Simulated and empirical labour, USA 1900-2000
3,5
empirical
normalised labour (1900=1)
3
simulated
2,5
2
1,5
1
0,5
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
2000
Capital accumulation feedback loop
<Gr oss
Output>
Capital
Inv e stme nt
Investment
Fraction A
Depre ciation
Inv e stme nt
Fraction
Depreciation
Rate A
Depre ciation
Rate
Investment
Fraction B
Depreciation
Rate B
Structural Shift
Time A
<Time>
Structural Shift
Time B
Parameters for USA 1900-2000
•
Investment Fraction A=0.081
• Depreciation Rate A=0.059
• Structural Shift Time A=1970
Investment Fraction B=0.074
Depreciation Rate B=0.106
Structural Shift Time B=1930
Capital investment and depreciation
Simulated investment and depreciation, USA 1900-2000
1.8
1.6
investment
depreciation
normalised capital (1900=1)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
Capital – validation by empirical fit
Simulated and empirical capital, USA 1900-2000
14
empirical
normalised capital (1900=1)
12
simulated
10
8
6
4
2
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
Output – validation of full model, US 1900-2000
Simulated and empirical capital, USA 1900-2000
14
empirical
normalised capital (1900=1)
12
simulated
10
8
6
4
2
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
LINEX fits for GDP, Japan and US
1900-2000.
Empirical and estimated GDP (using LINEX),
US and Japan 1900-2000
8000
empirical GDP, Japan
7000
GDP (thousand billion 1992$)
predicted GDP, Japan
6000
5000
empirical GDP, US
predicted GDP, US
4000
3000
2000
1000
0
1900
1920
1940
1960
year
1980
Estimates of GDP, France 1960-2000
4
3.5
output (1960=1)
3
Y
LINEX
Time Dependent CD
Time Average CD
2.5
2
1.5
1
0.5
0
1963
1968
1973
1978
1983
1988
1993
A commonly used reference mode
Energy Intensity of Capital, USA 1900-2000.
28
Start of the Great Depression
b/k - total primary exergy supply
(energy carriers, metals,
minerals and phytomass exergy)
26
24
e/k - total fuel exergy supply
(energy carriers only)
22
index
20
18
16
14
12
10
End of World War II
8
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
2000
The REXS alternative
Simulated and empirical primary exergy intensity of output,
USA 1900-2000
1.2
1
Average rate of
decline 1.2%
per annum
r/y (1900=1)
0.8
0.6
0.4
0.2
empirical
simulated
0
1900
1910
1920
1930
1940
1950
year
1960
1970
1980
1990
The “dematerialising” dynamics
Declining resource
inte nsity of output
cumulative
output
experience
Continuing historical
tre nds of te chnical
e fficie ncy growth
Economic
output
cumulative exergy
pr oduction
experience
Use ful wor k
supply
Primary exergy intensity (B/GDP) of output
decay feedback mechanism.
<Gr oss
Output>
Primary Exe rgy
Inte nsity of Output
Rate of Decay
Fractional
Decay Rate
Primary Exe rgy
Demand
Parameters
•
•
Product
Improvement
R&D Substitution of
Knowledge for Labour;
Capital; and Exergy
Rate of Decay = Fractional
Process
Decay Rate*Primary Exergy
Improvement
Intensity of Output
Lower Limits to
Costs of
Fractional Decay Rate=0.012 INCREASED REVENUES
To the right:
Increased Demand for
Final Goods and Services
Processes aggregated in
the REXS dynamics
Substitution of
Exergy for Labour
and Capital
Production
Economies of
Scale
Lower Prices of
Materials &
Energy
Projections of future output
Altering the future rates of the energy intensity of
output
•The average decay rate of the exergy intensity of
output (R/GDP) for the period 1900-1998 is 1.2%
•The simulations involved increasing or
decreasing this parameter from 1998 onwards,
while keeping the values of all other parameters
fixed.
•The following illustrations provide a summary of
the results.
Varying rates of dematerialisation
Primary Exergy Intensity of Output Decline Rate
0
historical trend
50%
75%
95%
100%
The constant
rate of exergy
intensity decline
was altered to
-0.5
vary between –
0.55 and –1.65
% p.a.
(%) -1
-1.5
-2
1900
1938
1975
Year
2013
2050
Effects on ‘efficiency’ improvements
If technical
efficiency does
not increase in
pace with ‘dematerialisation’
The rate of growth
slows.
0.3
efficiency
The ‘business as
usual’ case:
Technical Efficiency of Primary Exergy Conversion
0.4
historical data
50%
75%
95%
100%
0.2
0.1
0
1900
1938
1975
Year
2013
2050
GDP forecasts “dematerialisation
scenarios” ,US 2000-2050
95%
100%
150
Index (1900=1)
The sensitivity of
future projections
of GDP were
assessed, the red
line indicates the
‘business as usual’
for a fractional
decay rate of
energy intensity of
output –1.2 % per
annum and
technical efficiency
at 1% p.a.
Gross Output
200
historical data
50%
75%
100
50
0
1900
1938
1975
Year
2013
2050
Historical and forecast GDP for alternative
15. Historicalof
and the
forecast energy
GDP, for alternative
rates of
rates of Figure
decline
intensity
of
decline of energy intensity of output, US 1950-2050
120
100
GDP (1900=1)
80
output, US 1900-2000
1.2% per annum
1.3% per annum
1.4% per annum
1.5% per annum
empirical
60
40
20
0
1950
1975
2000
year
2025
2050
Forecast GDP growth rates for three alternative
technology scenarios (US 2050).
Alternative Technology Scenarios
Low
Growth rate f
GDP
Mid
f
GDP
High
f
GDP
Minimum
0.16% -2.97%
0.43% -1.89% 1.11%
1.94%
Average
0.40% -1.29%
0.72% 0.38% 1.18%
2.20%
Maximum
0.62% 0.92%
0.89% 1.75% 1.23%
2.63%
Note the feedback between f growth and GDP growth
Historical
and
forecast
technical
efficiency
Figure
13. Historical
and forecast
technical efficiency
of energy of energy
conversion, for 3 alternative rates of technical efficiency growth,
conversion,
for 3 alternative
rates of technical
US 1950-2050
efficiency growth, US 1950-2000.
0.35
technical efficiency (f)
0.3
0.25
low
mid
high
empirical
0.2
0.15
0.1
0.05
0
1950
1975
2000
year
2025
2050
Historical and forecast GDP, for 3 alternative
Figure 14. Historical and forecast GDP, for 3 alternative rates of
rates of technical
efficiency
growth, US 1950technical efficiency
growth, US 1950-2050
2050
70
60
GDP (1900=1)
50
low
mid
high
empirical
40
30
20
10
0
1950
1975
2000
year
2025
2050
Conclusions
• Travail utile comme facteur de production
• Application du 2° loi pour « proxy » de progrès
technologique
• Fonction LINEX et représentation Systèmes
Dynamique permettant
– Estimation historique
– « substitution dynamique » suite aux progrès
– Feedback entre progrès technologique et le quantité
et qualité des sources énergétique et l’efficacité
d’utilisation