1surendercarbon sensitive productivity growth, climate
TRANSCRIPT
Carbon Sensitive Productivity Growth, Climate Change and
Institutions
Surender Kumar
Department of Business Economics, University of Delhi, New Delhi
Motivation
• Concerns about climate change impacts and at the same time, concerns about the costs of abating emissions have made climate change an issue of central concern
• Climate change impacts occurs through affecting the productivity of resources
• Mitigation of CO2 emissions diverts resources from the conventional usage
• Better institutions helps in better utilization of resources and supposed to improve productivity and thus lowers the impacts of climate change and cost of mitigation
Related Literature
• Efficiency and productivity measurement under strong and weak disposability of CO2
emissions (e.g., Fare et al 1996, Murty and Kumar 2002, Kumar and
Khanna 2009, Kumar and Managi 2009).
• Climate change impacts: sectoral studies, IAMs, econometric studies (e.g., Tol 2002, Deschenes and Moretti 2007, Nordhaus 2010, Dell et al. 2008)
• Cost of abatement: CGE/IAMs models (e.g., Nordhaus, 2011)
• Institutions and economic growth (e.g., Acemoglu and his colleagues 2001, 2002, 2003, 2005)
Research Questions
• Does CO2 emissions abatement lower the growth potential?
• Does climate change affect productivity growth
• Do better institutions helps in lowering the impacts of climate change
Outline
• Basic framework and production technology
• Directional output distance function and productivity measurement
• Productivity results
• Determinants of productivity: climatic factors and institutions
• Regression results
• Summary and conclusions
The basic framework
• y is desirable (good) output
• b is undesirable (bad) output – CO2 emissions
• x is the vector of inputs
• there are K countries
Technology
We model technology in terms of output sets P(x)P(x) = {(y, b): x can produce (y, b)}
Outputs are assumed to be weakly disposable
Desirable output is freely (or strongly) disposable
But bad output is not (if there is environmental regulation)
‘Goods’ are nulljoint with the ‘bads’
P(x) b)y,(imply 10 and )(),( ∈≤≤∈ θθθxPby
P(x) b),y(imply y y and )(),( ∈≤∈ ))xPby
0y then 0b and )(),( ==∈ xPbyif
Directional Output Distance Function
It seeks to increase the desirable outputs whilst simultaneously reducing the undesirable outputs, i.e.,
{ }( , , ; ) sup : ( , ) ( )oD g g Pβ β→
= + ∈y b x y b x
Graphical Presentation of Directional output distance function
Productivity growth, efficiency change and technical change
Productivity Change = Efficiency Change + Technical Change(Technological Diffusion) (Innovation)
Data
• 88 countries, 26 developed and 62 developing countries
• Time period: 1994‐2008
• Desired output: GDP (in 2005 PPP)
• Undesired output: CO2 emissions
• Inputs: labour, capital and energy consumption
Descriptive Statistics of Various Measures of Productivity in 2008 over 1994
Developed Countries Developing Countries
Variable MeanStd. Dev. Min Max Mean
Std. Dev. Min Max
EFFCHS 0.98 0.15 0.74 1.43 0.92 0.21 0.39 1.40TCS 1.22 0.17 1.00 1.73 1.26 0.18 1.04 1.89PCS 1.20 0.28 0.75 1.97 1.15 0.25 0.54 1.80EFFCHW 1.00 0.13 0.75 1.43 1.05 0.11 0.75 1.42TCW 1.22 0.15 1.02 1.69 1.06 0.17 0.90 1.83PCW 1.22 0.24 0.76 1.91 1.11 0.20 0.80 1.76
ENVEFF 0.96 0.06 0.77 1.00 0.66 0.24 0.12 1.00
Income and productivity change under strong disposability of CO2 emissions
TZA
ETH
MOZZMB
NGA
BEN
KEN
NPL
BGD
SEN
SDN
HTI
TGO
CMR
MNG
VNM
AGO
GHA
PAK
IND
BOL
NIC
HND
ZWE
CHN
ALB
LKA
IDN
MAR
ECU
GTM
PHL
JAM
JOR
LBN
SLV
PER
PRY
DZA
TUN
PAN
NAM
BGR
TUR
BWA
COL
THA
GAB
ROM
BRA
CRI
POL
MEXMYS
CHLURY
ZAF
HUN
ARG
MLT
GRC
PRT
BHR
ESP
CYP
ISR
NZL
IRL
FIN
SGP
GBR
SWE
ITA
FRA
NLD
AUS
AUTBEL
ISL
CAN
DNK
JPN
NOR
CHE
USA LUX
.4.6
.81
1.2
1.4
Effic
ienc
y C
hang
e
0 10000 20000 30000Per Capita GDP (1995)
TZA
ETH
MOZ
ZMBNGA
BEN
KEN
NPL
BGD
SEN
SDN
HTI
TGO
CMR
MNG
VNM
AGO
GHA
PAK
IND
BOL
NIC
HND
ZWE
CHNALB
LKAIDN
MAR
ECU
GTM
PHL
JAM
JORLBN
SLV
PER
PRYDZA
TUN
PAN
NAM
BGRTUR
BWA
COL
THAGAB
ROM
BRA
CRI
POL
MEX
MYS
CHL
URY
ZAF
HUN
ARG
MLT
GRCPRT
BHR
ESP
CYP
ISR
NZL
IRL
FIN
SGP
GBR
SWE
ITA
FRANLD
AUS
AUT
BELISL
CAN
DNKJPN NOR
CHE
USA
LUX
11.
21.
41.
61.
8Te
chni
cal C
hang
e
0 10000 20000 30000Per Capita GDP (1995)
TZA
ETH
MOZ
ZMB
NGABEN
KENNPLBGDSEN
SDN
HTITGO
CMR
MNG
VNM
AGO
GHA
PAK
IND
BOL
NICHND
ZWE
CHN
ALB
LKA
IDNMAR
ECU
GTM
PHL
JAM
JOR
LBN
SLV
PER
PRY
DZA
TUN
PAN
NAM
BGR
TUR
BWA
COL
THA
GAB
ROM
BRA
CRI
POLMEX
MYS
CHL
URY
ZAF
HUNARG
MLT
GRC
PRTBHR ESP
CYP
ISR
NZL
IRL
FIN
SGP
GBR
SWEITA
FRA
NLD
AUS
AUT
BELISL
CAN
DNK
JPN
NOR
CHE
USA
LUX
.51
1.5
2Pr
oduc
tivity
Cha
nge
0 10000 20000 30000Per Capita GDP (1995)
Income and productivity change under weak disposability of CO2 emissions
TZA
ETH
MOZ
ZMB
NGA
BEN
KEN
NPL
BGD
SEN
SDN
HTI
TGO
CMR
MNG
VNM
AGO
GHA
PAK
IND
BOL
NIC
HND
ZWE
CHNALB
LKA
IDNMAR
ECU
GTM
PHL
JAMJOR
LBN
SLV
PER
PRY
DZATUN
PAN
NAM
BGR
TUR
BWA
COL
THA
GAB
ROM
BRA
CRI
POL
MEXMYS
CHLURY
ZAF
HUN
ARG
MLT
GRC
PRT
BHR
ESP
CYP
ISRNZL
IRL
FIN
SGP
GBR
SWE
ITA
FRANLD
AUS
AUT
BEL
ISL
CAN
DNKJPN NOR
CHE USA LUX
.81
1.2
1.4
Effi
cien
cy C
hang
e
0 10000 20000 30000Per Capita GDP (1995)
TZA
ETH
MOZ
ZMB
NGA
BENKEN
NPL
BGD
SEN
SDN
HTI
TGO
CMRMNG
VNM
AGOGHA
PAKIND
BOLNIC
HND
ZWECHNALB
LKA
IDNMARECU
GTM
PHL
JAMJOR
LBN
SLV
PER
PRY
DZATUNPAN
NAM
BGR
TUR
BWA
COL
THA
GAB
ROM
BRA
CRI
POL
MEX
MYS
CHL
URY
ZAF
HUN
ARG
MLT
GRC
PRT
BHR
ESP
CYP
ISR
NZL
IRL
FIN
SGP
GBR
SWEITA
FRANLD
AUS
AUT
BEL
ISL
CAN
DNK
JPN
NORCHE
USA
LUX
.81
1.2
1.4
1.6
1.8
Tech
nica
l Cha
nge
0 10000 20000 30000Per Capita GDP (1995)
TZAETH
MOZ
ZMB
NGA
BEN
KEN
NPL
BGDSEN
SDN
HTI
TGO
CMR
MNG
VNM
AGO
GHA
PAK
IND
BOL
NIC
HND
ZWE
CHNALBLKA
IDNMAR
ECU
GTM
PHL
JAMJOR
LBN
SLV
PER
PRY
DZATUN
PAN
NAM
BGR
TUR
BWACOL
THA
GAB
ROM
BRA
CRI
POL
MEX
MYSCHL
URY
ZAF
HUN
ARG
MLT
GRC
PRT
BHR
ESP
CYPISR
NZL
IRL
FIN
SGP
GBRSWE
ITAFRANLD
AUS
AUT
BEL
ISL
CAN
DNK
JPN
NOR
CHE
USA
LUX
.51
1.5
2Pr
oduc
tivity
Cha
nge
0 10000 20000 30000Per Capita GDP (1995)
Productivity Change results
• Productivity growth happened both in developed and developing countries both under strong disposability and weak disposability of CO2 emissions.
• In developed countries, there is no efficiency improvement and growth in productivity is led by innovations.
• In developing countries growth in productivity is led by innovations under strong disposability but under weak disposability, it is function of both innovations and diffusion of technologies.
• Restrictions on the disposal of CO2 emissions is costly both in developed and developing countries, but it impacts much the growth potential of developing countries, though it is cheap to reduce emissions in these countries
Productivity change in major economies
0.000
0.200
0.400
0.600
0.800
1.000
1.200
1.400
1.600
EFFCHs TCs PCs EFFCHw TCw PCw ENVEFF
CHNINDJPNUSAOECD
Temperature and productivity change under strong disposability of CO2 emissions
CAN
MNG
NOR
ISL
FIN
SWE
CHN
CHE
USA
AUT
CHL
POL
DNK
GBRLUX
ROM
IRLNLD
BEL
HUN
BGR
NZLFRA
TURJPN
ITA
ALB
NPL
ARG
ESP
GRC
PRT
LBN
URY
CYPZAF
ISR
MLT
JOR
MAR
PAK
NAM
TUNPER
IRQ
BOL
MEX
ZMB
AUS
ZWE
AGO
BWA
TZA
ECU
DZA
IND
VNM
PRY
ETH
GTM
MOZSLVGAB
KEN
CRI
BRABGD
HTI
CMR
COL
MYS
PAN
HNDIDN
PHL
JAM
THA
OMN
BHR
NIC
SGP
TGONGA
GHA
SDN
BEN
LKA
SEN
.4.6
.81
1.2
1.4
Effic
ienc
y C
hang
e
-10 0 10 20 30Temp(average over 1980-2008)
CAN
MNG
NOR
ISL
FINSWE
CHN
CHE
USAAUT
CHL
POL
DNK
GBR
LUX
ROM
IRL
NLD
BEL
HUN
BGR
NZL
FRA
TURJPN
ITAALB
NPL
ARG
ESPGRCPRT
LBN
URY
CYP
ZAF
ISR
MLT
JOR
MAR
PAK
NAMTUN
PER
IRQ
BOL
MEX
ZMB
AUS
ZWE
AGO
BWA
TZA
ECU
DZA
INDVNM
PRY
ETH
GTM
MOZ
SLV
GAB
KEN
CRI
BRA
BGD
HTICMR
COL
MYSPANHNDIDN
PHL
JAM
THA
OMNBHR
NIC
SGP
TGO
NGA
GHA
SDN
BEN
LKA
SEN
11.
21.
41.
61.
8Te
chni
cal C
hang
e
-10 0 10 20 30Temp(average over 1980-2008)
CANMNG
NORISL
FIN
SWE
CHN
CHE
USA
AUT
CHL
POLDNK
GBR
LUX
ROM
IRL
NLD
BELHUN
BGR
NZL
FRA
TURJPN
ITA
ALB
NPL
ARGESP
GRC
PRT
LBN
URY
CYP
ZAFISR
MLT
JOR
MAR
PAK
NAM
TUN
PER
IRQ
BOLMEX
ZMB
AUS
ZWE
AGOBWA
TZA
ECU
DZA
IND
VNM
PRYETH
GTM
MOZ
SLV
GAB
KEN
CRI
BRABGD
HTI
CMR
COL
MYS
PAN
HNDIDN
PHL
JAM
THA
OMN
BHR
NIC
SGP
TGO
NGA
GHA
SDNBEN
LKA
SEN
.51
1.5
2Pro
duct
ivity
Cha
nge
-10 0 10 20 30Temp(average over 1980-2008)
Temperature and productivity change under weak disposability of CO2 emissions
CAN
MNG
NOR
ISL
FIN
SWE
CHN
CHEUSAAUT
CHL
POL
DNK
GBRLUX
ROM
IRLNLD
BEL
HUN
BGR
NZL
FRA
TURJPN
ITA
ALB
NPL
ARG
ESP
GRC
PRT
LBN
URY
CYPZAF
ISR
MLT
JORMAR
PAK
NAM
TUN
PER
IRQ
BOL
MEX
ZMB
AUS
ZWE
AGOBWA
TZA
ECU
DZA
IND
VNM
PRY
ETH
GTM
MOZ
SLV
GAB
KEN
CRI
BRABGD
HTI
CMRCOL
MYS
PAN
HND
IDN
PHL
JAMTHA
OMN
BHR
NIC
SGP
TGO
NGA
GHASDN
BEN
LKA
SEN
.81
1.2
1.4
Effic
ienc
y C
hang
e
-10 0 10 20 30Temp(average over 1980-2008)
CAN
MNG
NOR
ISL
FIN
SWE
CHN
CHE
USA
AUT
CHLPOL
DNK
GBR
LUX
ROM
IRL
NLD
BEL
HUN
BGR
NZL
FRA
TURJPN
ITA
ALB
NPL
ARG
ESP
GRC
PRTLBN
URY
CYP
ZAF
ISR
MLT
JOR
MAR
PAKNAM
TUN
PER
IRQBOL
MEXZMB
AUS
ZWEAGO
BWATZA
ECUDZA
INDVNMPRYETH
GTM
MOZ
SLV
GAB
KEN
CRI
BRA
BGD
HTI
CMRCOL
MYSPAN
HND
IDN
PHL
JAMTHA
OMNBHR
NIC
SGP
TGO
NGA
GHA
SDN
BEN
LKASEN
.81
1.2
1.4
1.6
1.8
Tech
nica
l Cha
nge
-10 0 10 20 30Temp(average over 1980-2008)
CAN
MNG
NOR
ISLFIN
SWE
CHN
CHE
USA
AUT
CHL
POLDNK
GBR
LUX
ROM
IRL
NLDBEL
HUNBGR
NZL
FRATURJPN
ITAALB
NPL
ARGESP
GRC
PRT
LBN
URY
CYP
ZAF
ISR
MLT
JOR
MARPAK
NAM
TUN
PER
IRQ
BOL
MEX
ZMB
AUS
ZWE
AGO
BWA
TZAECU
DZAIND
VNM
PRY
ETHGTM
MOZ
SLV
GAB
KEN
CRI
BRA
BGD
HTI
CMRCOL
MYS
PAN
HND
IDN
PHL
JAMTHA
OMN
BHR
NIC
SGP
TGO
NGA
GHA
SDN
BEN
LKA
SEN
.51
1.5
2Pr
oduc
tivity
Cha
nge
-10 0 10 20 30Temp(average over 1980-2008)
Productivity Growth and Climate Change
• Negative relationship between temperature and productivity change under both scenarios
• Under strong disposability, the increase in temperature levels lowers the technological adoption which is not compensated by the innovations.
• When the CO2 emissions disposal is restricted, innovations becomes costly though it is possible that at some higher temperature levels the adoption may improve
Determinants of Productivity and its components
• Cumulative Productivity (CP) and both of its components are regressed on the initial level of productivity/efficiency, level of temperature, soil quality, AVEXPR and SDFORMALISM index
• Regressions are run for all the sample countries, and separately for developed and developing countries
• Relationship between CP or its components and initial levels of productivity/efficiency shows convergence/ divergence across countries.
• +ve relation between CP and initial productivity shows divergence across countries
• ‐ve relation between CP components and initial inefficiency indicates divergence across countries
• Climate change is expected to lowers the productivity growth, i.e., a negative relationship between PC and temperature
• Better property rights are expected to positively influence the growth of productivity, i.e., a positive association between PC and AVEXPR
• More complex is the system, less productivity growth is expected, i.e., a negative relationship between SDFORMALISM and PC
Descriptive Statistics of Variables used in Regressions
Developed Countries Developing Countrie
Variable Obs MeanStd. Dev. Min Max Obs Mean
Std. Dev. Min Max
Soil Quality 26 48.50 34.07 0 139.7 62 79.37 40.9 12.9 182.7
Temperature 26 10.36 7.20 -7.08 26.6 62 21.10 6.23 0.24 28.2
AVEXPR 25 9.48 0.67 7.23 10 57 6.55 1.28 1.64 9SDFORMALISM 26 3.09 0.80 1.58 5.25 53 4.08 1.01 1.68 5.91
Regression Results: Productivity Change • A +ve association is found between CP and its initial level
which helps in explaining the growing inequality between countries. Divergence is happening even with in the developing countries under both the scenarios
• Soil quality levels impacts productivity growth in developed countries whereas the temperature levels affects in developing countries
• A ‐ve association between temperature level and CP level for the sample of developing countries under strong disposability indicates that climate change impacts are more pronounced in the developing countries.
• Better property rights and less complex systems helps in increasing the productivity level in developed countries under both the scenarios.
Regression Results: Efficiency Change
• Under strong disposability, neither divergence nor convergence is observed in either of the group.
• However, if the disposal of emissions is restricted then there is convergence in technological adoption across countries and within the groups of developed and developing countries.
• When the disposal of CO2 emissions is free, the temperature increases lower the technological diffusion in developing countries, but better property rights helps in lowering the impact of climate change and spread the diffusion
• Soil quality is not related to technological diffusion
• Less complex systems improves the diffusion in developed countries
• The process of the diffusion is negatively affected in developed countries due to climate change under weak disposability of the emissions
• But better property rights in developed countries helps in improving the process of technological diffusion
Regression Results: Innovations
• Under strong disposability there is divergence in innovation in developed countries but convergence in developing countries
• But if the disposal is made costly, divergence in innovation is common in both the groups
• Better property rights helps in improving the innovations in developed countries under weak disposability and in developing countries under strong disposability
• However, innovations are not related to temperature changes under either of the scenarios
Summary and Conclusions• Productivity growth is happening, and the divergence in
productivity growth helps in explaining the growing inequalitiesacross countries.
• Though there is higher potential in reducing CO2 emissions in developing countries at lower cost, but the costly disposable ofthe emissions lowers the growth potential of these countries.
• Climate change lower the productivity growth in developing counties through lowering the process of technological adoption
• Better property rights and governance systems results in higher productivity .
• Under strong disposability, better property rights helps in lowering the impacts on climate change on developing countries through improving the process of technological adoption.
