Transcript Titel des Deliverables by Andreas Löschel, Simon Koesler and

Structural Change and the Environment
Structural Change and the Environment:
An Application to the WIOD Database
Deliverable 7.2
by Andreas Löschel, Sascha Rexhäuser and Michael Schymura
WIOD Consortium Meeting, Seville,
5
25th – 28th June 2011
Structural Change and the Environment
Content
1
Introduction
2
Data and Econometric Model
3
Endogenity Problems
4
Instruments
Results
5
Results
6
Conclusions and Drawbacks
2
Structural Change and the Environment
1
Introduction
Research Questions:

Is Free Trade Good or Bad for the Environment?

Which factors are affected by trade?

Direct or indirect impacts?
Methodology:

Structural Decomposition Analysis (SDA)

Panel Econometrics

Approach based on the seminal contribution by Antweiler,
Copeland and Taylor (2001, A.E.R.)
3
Structural Change and the Environment
2
Data and Econometric Model
Used data from WIOD:

Information about Gross Output, Share of high-skilled workers

Environmental Satellite Accounts

Trade Data
Other data sources:

Income, Population, Investments (PWT 6.3)

Depreciation, TFP growth, Capital Stocks (EPWT 3.0)

Geographical informations such as distance, area etc. (CIA)

Data on patents (several sources)
4
Structural Change and the Environment
1
2
Data and Econometric Model
Econometeric Implementation:
∆ POLLUTION = Scale + Composition + Technique
SO2 _ Emissionenit  0  1INCit  2 INCit 
2
 3KLit  0 TOit  it
 1TOit  REL.KLit  2 TOit  REL.KLit 
2
 3 TOit  REL.INCit  4 TOit  REL.INCit 
2
 5 TOit  REL.KLit  REL.INCit
 1HELSINKIit  2OSLOit  it
see Antweiler, Copeland, and Taylor (2001, A.E.R)
5
Structural Change and the Environment
1
2
3
Endogenity Problems
Problems:

Trade (TO) is endogenous

Income (INC) is endogenous too

Regulation (Helsinki)?
Reasons for Endogenity (among others):

Measurement Errors

Omitted Variables Problem

Simultaneity  TO, INC, HELSINKI
6
Structural Change and the Environment
1
2
3
4
Instruments
Instrument Trade Openess (1/2):

Taken from the trade literature, Gravity-Model
(see Frankel und Romer (1999, A.E.R.))

e.g. bilateral distance, common border, etc.:
- strong correlation with actual trade openess
- not correlated with SO2-Emissions or the error term

2SLS  Stage 1:
ln(TOijt )   0  1 ln(Dis tanceijt )   2 ln(ActPopit )   3 ln(ActPop jt )
  4 ln(Areait  Area jt )   5 (LLit  LL jt )   6 CB   7 CC
  8 CBijt (LLit  LL jt )  ijt
7
Structural Change and the Environment
1
2
3
4
Instruments
Instrument für Trade Openess (2/2):
Actual vs. Constructed trade share
8
Structural Change and the Environment
1
2
3
4
Instruments
Instrument for Per Capita Income (1/2):

Taken from the growth literature:
(see Mankiw, Romer und Weil (1992, Q.J.E.) and Frankel and
Rose (2005, R.E.S.))

Conditional Convergence Hypothesis:
- Lagged income
- Investments and Depreciation
- Population growth, Growth of labor productivity
- additionaly from WIOD: share of high-skilled workers (HK)

2SLS  Stage 1:
 GDP 
 GDP 
 I 
ln 




ln


ln


0
1 
2 
  3 ln n  g   it  HKit  it
Pop
Pop
GDP

it

it

it 1
9
Structural Change and the Environment
1
2
3
4
Instruments
Instrument for Per Capita Income (2/2):
Actual vs. Constructed income per capita:
10
Structural Change and the Environment
1
2
3
4
5
Results (Endogenous)
11
Structural Change and the Environment
1
2
3
4
5
Results (Instruments)
IV vs. 2SLS
12
Structural Change and the Environment
1
2
3
4
5
6
Conclusions
Conclusions and drawbacks:

Currently it is hard to conclude whether free trade is good or bad
for the environment. But: postive trade induced composition
effects (in this case positive is bad)

Final report will include as many as possible WIOD data with
respect to the environmental satellites, the trade data, output
data, labor files and especially country coverage

Problem: Physical capital stock data is hardly available for the
time and country period covered by WIOD. One possible solution:
Perpetual Inventory Mehtod (Caselli (2005)).

Problem: Endogenity of Regulation
13
Structural Change and the Environment
Literature3
Antweiler, Werner, Brian R. Copeland und M. Scott Taylor (2001): Is
Free Trade Good for the Environment?, in: American Economic
Review, Vol. 91, No. 4, S. 877-908.
Caselli, Francesoco (2005): Accounting for Cross-Country Income
Differences, in: Handbook of Economic Growth, Vol. 1, No. 1,
Chapter 9, S. 679-741
Frankel, Jeffrey A. und David Romer (1999): Does Trade Cause
Growth?, in: American Economic Review, Vol. 89, No. 3, S. 379399.
Frankel, Jeffrey A. und Andrew Rose (2005): Is Trade Good or Bad
for the Environment?: Review of Economics and Statistics, Vol.
87, No. 1, S.85-91
14
14
Structural Change and the Environment
3
Managi, Shunsuke, Akira Hibiki und Tetsuya Tsurumi (2009): Does
Trade Openess Improve Environmental Quality, in: Journal of
Environmental Economics and Management, Vol. 58, No. 3, S.
346-363.
Mankiw, N. Gregory, David Romer und David N. Weil (1992): A
Contribution to the Empirics of Economic Growth, in: Quarterly
Journal of Economics, Vol. 107, No. 2, S. 407-437.
Staiger, Douglas und James H. Stock (1997): Instrumental Variable
Regression with Weak Instruments, in: Econometrica, Vol. 65, No.
3, S. 577-586.
15
15
Structural Change and the Environment
3
Thank you for your attention!
16
16