Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Environmental Kuznets Curve

VIEWS: 193 PAGES: 25

									• Topic Four: • The Environmental Kuznets Curve • The relationship between economic growth and environmental degradation across countries and within countries, across households and within households

1

Environmental Kuznets Curve with a turning point at “Y*”
pollution

Y* GNP Per-Capita
2

A statement of the EKC
• At the macro level, economic development first raises pollution problems but beyond some turning point, economic growth and pollution are negatively correlated • At the micro level, as a poor person grows richer she pollutes more but as she grows even richer her environmental impact declines
3

The Research Agenda
• Identify those environmental indictors for which the EKC might describe crosssectional and panel data • Measure the actual shapes of curves such as that presented in slide #1

4

Origins of This Hypothesis
• NAFTA and Free Trade
• Jagdish Bhagwati (2004) neatly summarizes this point. “The only value of these [EKC] examples is in their refutation of the simplistic notions that pollution will rise with income. They should not be used to argue that growth will automatically take care of pollution regardless of environmental policy. Grossman and Krueger told me that their finding of the bell-shaped curve had led to a huge demand for offprints of their article from antienvironmentalists who wanted to say that “natural forces” would take care of environmental degradation and that environmental regulation was unnecessary; the economists were somewhat aghast at this erroneous, ideological 5 interpretation of their research findings

Leading Economics Papers on this Subject
• • • • Grossman and Krueger (1995) Harbaugh, Levinson and Wilson (2002) Hilton and Levinson (1998) Confronting the EKC (2002)

6

Urban Lead Emissions
• Lead exposure lowers children’s IQ • An unintended consequence of increased driving in developing nations could be sharp increases in lead emissions

7

Hilton and Levinson
• • • • • • Factoring lead emissions for vehicles Lead emissions = gallons of gasoline * lead emissions per gallon Income growth increases gallons of gasoline Income growth reduces lead emissions per gallon The latter effect can dominate the former effect and this can generate a non-monotonic relationship between GDP and pollution
8

Factoring Pictures

9

Shifting Environmental Kuznets Curve
pollution
1980 2005

GNP Per-Capita
10

Relevance?
• The previous slide suggested that the EKC could shift “down” and “left” over time. • So what? • If the curve shifts down, this means that there is less pollution associated with any level of GNP per-capita • If the curve shifts left, this means that newly developing nations reach the “turning point” sooner!
11

What Factors could shift the EKC Down over time?
• Technological advance • Radical shifts in consumption patterns • Correct resource pricing --- the phase out of energy subsidies in ex-communist nations such as Hungry, Poland and the Czech Republic
12

Environmental Kuznets Curve Shifts for Ex-Communist Countries
pollution
1988 2005

Y* GNP Per-Capita
13

What Factors could shift the EKC’s Turning Point left over time?
• Environmental “consciousness” dynamics • Salient events --- why would this “consciousness and concern” change over time? • Bhopal, Chernobyl, 3 Mile Island, Love Canal • Al Gore’s book, Silent Spring, movies and social interactions • Technological innovation and the induced innovation hypothesis • Diffusion of “green ideas”
14

EKC without Government?
• A household gains utility from cooking, clothing and it doesn’t like pollution • Define C = cooking services • Define P = pollution • The household goes to the market and purchases fuels. Some fuels such as wood are cheaper than LPG but wood emits more pollution per unit.
15

World Bank Living Standards Measurement Surveys
• The LSMS surveys are available from the World Bank’s web page • Typical survey collects a representative sample of households (perhaps 10,000) • Asks a large number of questions about household income, sources of income and consumption patterns
16

The Energy Ladder
• Suppose that LPG is twice as expensive as Wood as a fuel to purchase • Suppose that LPG/gallon creates 80% less pollution than Wood • As a household grows richer, the income effect would say “consume more LPG” if avoiding pollution is a normal good • Especially likely if diminishing returns to cooking
17

Household Level EKC with a turning point at “Y*”
pollution
Quality effect dominates Quantity effect

Quantity dominates quality effect

Y* Household Income
18

Chaudurhi and Pfaff
• Pakistan data from LSMS in the 1990s • They document evidence of a household level EKC • Externality is “internalized” • If household’s cooking smoke went out of its windows, much less likely to substitute to higher quality fuels as the household grows richer • http://www.columbia.edu/~ap196/
19

1997 to 1999 California Random Roadside Emissions Data
• 25,000 vehicles randomly pulled off the roads • Vehicles’ emissions are measured • Each vehicle’s characteristics are recorded • The vehicle owner’s zip code of registration is collected • This last piece of information allows me to test some “social science” hypotheses
20

California Vehicle Emissions for rich and poor households

Sample Income Percentile whole <25th 25th-75th 75th-95th 95th+ hydrocarbon 108.023 141.840 112.153 75.956 54.790 carbon monox 0.771 1.053 0.788 0.540 0.447 nox 699.023 887.093 699.248 565.161 575.998 Fancy Vehicle 0.058 0.047 0.050 0.079 0.180 Average Miles per Year 9708.386 8706.437 9623.377 10639.240 10462.300 Light Truck 0.300 0.302 0.301 0.303 0.251 income 60592.510 37060.680 55064.630 85409.900 150299.700

21

Vehicle Emissions Regressions: E = controls + b*log(inc) + U
log(zip income) log(days IM test) Engine Size Fancy Vehicle USA Vehicle Light Truck Log(miles) Constant beta -77.82 11767.00 t-stat -11.64 beta -42.13 t-stat -6.66 -0.83 -5.25 4.02 1.85 2.85 6.71 beta -36.42 4.39 -2560.37 -32.51 12.91 11.31 4.33 447.59 Yes Yes Yes t-stat -5.61 2.61 -1.09 -3.90 2.93 2.63 1.66 5.83

953.22

5.84 -1797.24 -37.33 17.37 7.66 6.45 12.67 487.31 Yes Yes Yes

Climate Variables Included Yes Calendar Year Fixed Effects Yes Model Year Fixed Effects No

22

Inference?
• Per-mile of driving, rich people pollute less than poorer people • Composition effect --- rich own newer vehicles, more likely to own “fancy” vehicle • Technique effect– multivariate regression controls for many factors and yet income’s coefficient “b”<0 and statistically significant • Perhaps, richer people go to their mechanic more often than poorer people
23

Why are richer people’s vehicles polluting less?
• Not virtue • It just happens that a high quality well functioning vehicle does not emit much • The rich are likely to view their contribution to urban pollution as small • An unintended consequence of pursuing having a quality, functioning vehicle is low emissions • Don’t forget scale effects! The rich do drive more miles than the poor! (quantity versus quality of consumption)
24

Progress!
Hydrocarbons Oxides of Nitrogen 1.5 Carbon Monoxide

1

.5

0 1966 1971 1976 1981 1986 Model Year 1991 1996 2001

Predicted Vehicle Emissions by Model Year
25


								
To top