REBOUND EFFECT FOR PRIVATE TRANSPORT AND ENERGY SERVICES IN THE UK - PowerPoint PPT Presentation

REBOUND EFFECT FOR PRIVATE TRANSPORT AND ENERGY SERVICES IN THE UK IAEE European Conference, Vienna, September 3-6, 2017 Mona Chitnis (University of Surrey) Roger Fouquet (LSE) Steve Sorrell (University of Sussex)

Outline  Rebound mechanisms  Data overview  Model  Results  Discussion

Rebound mechanisms for households Total Rebound=Direct Rebound+Indirect Rebound Indirect Rebound Lower Lighting heating More or less emissions cost and fuel bills Less emissions Higher room temperature or leaving the heating on for longer Lower running Direct Rebound More emissions cost for heating

Measuring energy service price and Consumption Data Sources (e.g. BEIS, ONS) p E  E Energy Price Energy Consumption Energy Efficiency (e.g. pence (e.g. TWh for lighting) (e.g. lumen-hours per kWh) per kWh) S      p p / S E E Energy Service Price Energy Service Consumption (e.g. £ per lumen-hour) (e.g. millions of lumen-hours)

Energy services in this presentation  Lighting  Heating  Wet and cold appliances  Electronics and computing  Cooking  Car 5

Average efficiency by energy service in the UK 1964-2015 (index, 1964=100) 400% 1400% Lighting, Heating, Wet & cold, Cooking, Car 350% Electronics & computing (1964=100) 1200% 300% 1000% 250% (1964=100) 800% 200% 600% 150% 400% 100% 200% 50% 0% 0% 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Lighting Heating Wet & cold appliances Cooking Car transport Electronics & computing 6

160 140 Real price of energy 140 120 Lighting, Heating, We & cold, Cooking, Car Electronics & computing ( 1964=100) services (index, 1964=100) 120 100 100 (1964=100) 80 80 60 60 40 40 20 20 0 0 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 Lighting Heating Cooking Electronics & computing Wet & cold appliances Car transport 200 Lighting and Appliances, Cooking, Heating, Car 180 160 (1964=100) 140 120 Real price of energy for 100 services (index, 1964=100) 80 60 7 Lighting and Appliances Cooking Heating Car transport

450 7000 Lighting, Heating, Wet & cold, Cooking, Car Consumption of energy Electronics & computing (1964=100) 400 6000 services per equivalised 350 5000 300 person (index, 1964=100) (1964=100) 4000 250 200 3000 150 2000 100 1000 50 0 0 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Lighting Wet & cold appliances Cooking Heating Car transport Electronics & computing 450 1000 Lighting, Wet & cold, Cooking, Heating, Car 900 400 Electronics & computing (1964=100) 800 350 700 300 600 (1964=100) 250 500 200 400 Energy consumption per 150 300 100 200 equivalised person (index, 50 100 1964=100) 0 0 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Lighting Wet & cold appliances Heating 8 Cooking Car transport Electronics & computing

Total rebound estimation (in terms of CO2)       ( s ) ( i ) Total Rebound = p i p S S  i ( i s ) where: x u w Direct Indirect i i  i  x u w s s  ( s ) : own-price elasticity of service s p  S : cross-price elasticity of service i with respect to service s ( i ) p S u i : CO2 intensity of service i u s : CO2 intensity of service s w i : budget share of service i w s : budget share of service s

Two stage budgeting model Household expenditure Stage 1 Other goods and Energy services Transport services Lighting Car Stage 2 Heating Other transport Wet and cold appliances Electronics and computing Cooking

Almost Ideal Demand System (AIDS) Stage 1:               r, z: 1, 2, 3 w ln p ln( x / P ) w t rt r ij zt r t t rz z r t  1 t z z where: w r =budget share of category r P z =price of category z x=total expenditure per equivalised person   P=Laspeyres-like price index L 0 ln P w ln p t r rt r              Adding up: 1 , 0 , 0 , 0 r r rz rz r r r r   rz   Symmetry: rz    0  Homogeneity: rz r

Almost Ideal Demand System (AIDS) Stage 2:               r w ln p ln( x / P ) w t i, j: 1, …, 4 it i ij jt i rt rt ij j i t  1 t   j r j r where: w i =budget share of service i P j =price of service j x r =total expenditure on services per equivalised person   L 0 ln P w ln p P r =Laspeyres-like price index rt it i i             1 , 0 , 0 , 0  Adding up: i i ij ij i i i 1   ij   Symmetry: ij    0  Homogeneity: ij i

Elasticities for two-stage budgeting model ■ Combined elasticities of stages 1 and 2 to obtain total price and expenditure elasticities for each energy and transport service (Edgerton 1997). Within group elasticities:   : Kronecker’s delta Income elasticity  1  r E rs r w equal to one when r    w r=s and zero    Uncompensated price elasticity rz r z e rz rz elsewhere. w r Total/between group elasticities: E  E E Income elasticity i ( r ) i r      e e E w [ e ] Uncompensated price elasticity ij rz ( r ) ij ( r ) i ( z ) j rz ( r )( z )  UK household annual time series data 1964-2015  Iterative Seemingly Unrelated Regressions (ISUR) method for system estimation

Estimated average elasticities Total expenditure elasticities for energy services Wet & cold Electronics & Lighting Heating Cooking Car appliances computing Expenditure 0.96 1.29 1.10 0.99 0.81 0.89 elasticity Total price elasticities for energy services Price elasticity Lighting Heating Wet & cold Electronics & Cooking Car appliances computing Lighting -0.936 0.381 0.088 0.059 0.036 -0.037 Heating 0.040 -0.696 0.065 0.031 0.044 -0.050 Wet & cold 0.050 0.343 -0.922 0.044 0.049 -0.043 appliances Electronics & 0.061 0.438 0.098 -0.939 0.064 -0.039 computing Cooking 0.037 0.382 0.084 0.048 -0.917 -0.032 Car -0.004 -0.03 -0.006 -0.003 -0.004 -0.54

Estimated average rebound effects Rebound effects between energy and transport services Lighting Heating Wet & cold Electronics Cooking Car appliances & computing Lighting 93.6% -39.4% -8.0% -5.1% -3.7% 1.8% Heating -3.9% 69.6% -5.7% -3.8% -3.9% 1.4% Wet & cold appliances -5.5% -39.3% 92.2% -5.2% -5.2% 1.8% Electronics & -7.0% -35.6% -8.3% 93.9% -5.7% 2.0% computing Cooking -3.6% -42.7% -8.0% -5.5% 91.7% 1.9% Car 0.7% 9.7% 1.4% 0.7% 0.6% 54.2% Direct, indirect and total rebound effects for energy and transport services Direct Indirect rebound Total rebound rebound (energy and transport services only) Lighting 93.6% -54.3% 39.31% Heating 69.6% -15.9% 53.68% Wet & cold appliances 92.2% -53.3% 38.92% Electronics & computing 94.0% -54.5% 39.40% Cooking 91.7% -57.8% 33.91% Car 54.2% 13.1% 67.26%

Discussion  Data assumptions and limitations  No backfire, but rebound is not negligible and should not be neglected  Indirect rebound plays an important role  Direct rebound is insensitive to the estimation method/variables in the AIDS model  Indirect rebound of non-energy goods is neglected  Average rebound over the sample period  rebound over the sample period  Comparison with rebound based on energy/transport fuel demand

REBOUND EFFECT FOR PRIVATE TRANSPORT AND ENERGY SERVICES IN THE UK IAEE European Conference, Vienna, September 3-6, 2017 Mona Chitnis (University of Surrey) Roger Fouquet (LSE) Steve Sorrell (University of Sussex)