Transcript Environmental Kuznets Curves (EKC)

Issues
• Sustainability and the Environmental
Kuznets curve hypothesis
• EKC  (environmental) Innovation
• Global issues: Climate change policies 
• The Kyoto protocol and environmental
policy issues
• The EU emission trading experience
Sustainability and the
Environmental Kuznets Curves
(EKC)
Key issues
• Sustainable development
– Weak and strong economic perspectives
– Daly idea of zero growth
– Renewable and non renewable resources
• Environmental efficiency of economic systems
– The critical role of innovation only engine for
sustainaed economic growth and sustainability
(Solow models)
– Sustanaibility and sustaianed growth entangled
issues: trade offs but also complementarities through
the role of innovation
What is sustainable development?
• SD is the achievement of a sustained path of
economic growth which does not undermine
future generation possibilities of consumption
• We may define what “future generation” means
– An orthodox economist would claim that this depends
on our time preference  discount rate reasoning..
– The higher the discount rate, depending on
consumption and oportunity costs factors, the less
future benefits and costs are valued…
– r = pure time myopic preference + consumption
growth; otherwise equals tha market oppotunity cost,
the foregone benefit of an investment
• Recall that GNP=C+I
• Recall that NetNP= GNP – depreciation of
capital
• Capital stocks dynamics depends on
accumulation and depreciation
SD is linked to Total capital or
natural capital?
• Total capital = manmade + human capital +
natural capital
• Each capital stock is defined by a rate of growth,
I – Deprec.
• If I=dep, then capital is steady
• Y=(TOT-K)
• Thus, a first intuitive golden rule for SD is that
total K should be at least constant, Inv should at
least match depreciation..
• Genuine saving rule: INV >= depreciation
..but..
• This may imply a decreasing natural
capital stock, if natK is substituted by other
forms
– This is the western country history
– i.e. arab countries management of non
renewable resources
– UK oil exploitation
– In any case, rents from natural resource use
should be re-invested..
• Thus, weak sustainability may also imply a complete
exhaustion of natural resources…
• Strong sustainability is instead stressing the critical role
of some natural capital forms…irreversible losses…ecosystemic losses
• The genuine saving rule is applied to specific
environmental assets
– i.e. compensation projects
– It works for renewable resources (forests, fishery..)
– Striking difference between the management of non renew
resources (the problem is a correct price and a path of
exploitation which takes into account the existence of an
alternative backsyop technology) and renewable resources,
which often posses use and non use values…
• SD is also possible in case a reduced amount of capital is inherited
by future generations….
• …but this capital must be more productive..more efficient..
• We go back to the role of environmental innovation in triggering
higher resource efficiency of the economy
• A key issue is what he driving forces of innovation are:
– Prices (neoclassic view)
– Policy  which kind of policy…static reasoning demonstrate the higher
efficiency of green taxes and tradable permits (over CAC)…dynamic
efficiency should also be higher for economic instruments, but it is more
an empirical matter
– Firm internal strategies..Porter hp..firm gains from innovation in the long
run, to achieve new competitive advantages…hp at macro and micro
level
– A weak version of the hp claims that in the long run the policy costs are
lower than the induced innovation gains…NET benefits..
..summing up..
• SD depends on the decision on how much investing in
each period…(recall Y=C+I)..a part of the investment is
in innovation (tech and organizational)
• ..but even sustained economic growth (Solow Model) is
possible only in presence of technological change
enhancing factor productivity..
• SD intrinsically depends on innovation, which is an
investment, which also depends on economic growth..
• The possibility of achieving a SD path relies on the
extent to which innovation investments are capable of
reducing the impact of a sustained economic growth..
• This issue is known as Delinking: environmental impact
from economic growth
Delinking
• Advanced economic systems have been
characterised by a decreasing intensity of
energy and materials per unit of output, driven
by technological dynamics and regulatory
pressures. OECD recently published a
document (OECD, 2002) containing an updated
evidence on the de-linking from economic
growth, concerning diverse environmental
indicators, such as climate change, air pollution,
water quality, waste management, and material
use.
• Delinking may occur on a relative basis (the
elasticity of the environmental impact indicator
with respect to an economic driver is positive,
but less than unity) or on an absolute basis
(negative elasticity).
• The assessment of both de-linking processes
can be referred to the mostly applied research
field concerning Environmental Kuznets Curves
(EKC).
• The hypothesis derives from the original
analysis of Kuznets on the relationship between
income level and income distribution
• The EKC hypothesis is shortly that for many environmental impacts,
an inverted U-shaped relationships between per capita income and
pollution is documented.
• The concentration of a certain pollutant first increases with
income/production, reflecting a scale effect, more or less
proportional, then eventually starts to decrease, de-linking from
income even on an absolute basis.
• More specifically, the hypothesis predicts that the “environmental
income elasticity” decreases monotonically with income, and that it
eventually changes its sign from positive to negative, thus defining a
turning point for the inverted U-shaped relationship.
• It does not derive from a theoretical model, it is an intuitive
conceptual approach, inductive in nature..though some theoretical
explanations have emerged…
EKC motivations
•
Supply side
– Technology driven by economic growth (profits and investments..)
– The share of cleaner activities in GDP increases with the scale of the economy
(scale + composition effects)
– As scarcity increases, market prices should reflect it..self-regulatory mechanism?
– Environmental policy more likely in a developed economy  economic and
political conditions needed
– Property right enforcement (policy issue)
•
Demand side
– Environmental quality is a normal luxury good (as culture)..higher incomes mean
higher WTP for the environmental services..higher taxes are possible, new
markets are profitable..
– Preferences change as the society develops..the marginal value of consumption
is positive but decreasing
– Environmental costs are increasing even steeply…growth benefits
decreasing….even a simple marginal cost-benefit scheme may explain why
delinking may occur
•
As it is evident, many forces play their role, in the interplay between supply
and demand, and between policy and spontaneous market dynamics
Policy relevance
• The EKC evidence may support the idea
that no policy is needed…market forces
and market dynamics are self-sufficient in
inverting the income-environment link
• BUT the environmental impact may be
higher than what is defined as
sustainable…policy efforts are needed to
support and correct markets..affecting the
shape of the EKC
• Empirical evidence, which has mainly
concerned air emissions, is still
ambiguous. Some pollutants show a
turning point, though it shared view that
some critical externalities, like CO2 and
waste flows, are monotonically rising with
income. At best, relative de-linking may be
occurring (Stern, 2004).
• Air quality indicators
– Local air quality (CO, sulfur, PM) seem to show an
inverted U-shape with income.
– Global pollutants either rise monotonically with
income or eventually present very high turning point
(not reached if not by US)
– * private/public goods as far as countries are
concerned..free riding on global commons  policy
needed
• Water indicators
– The turning point is generally higher
– EKC for some indicators (local issues)
– N shape? (Borghesi, 1999)
waste
• Empirical evidence on Delinking
concerning environmental waste indicators
is probably the scarcest. Contributions
providing results for waste are rare. Cole
et al. (1997) find no evidence for an
inverted U-shape EKC curve concerning
municipal waste
• See the Mazzanti-Zoboli paper (2005) on
waste and delinking…
• There is currently no evidence concerning
de-linking with respect to primary sources
of waste in Europe (i.e. municipal and
packaging waste), which have been
targeted by waste-oriented European
Directives aimed at reducing diverse
environmental externalities associated to
waste production and disposal
Some evidence
• Turning points 2003$/per capita
(international studies)
– CO2:
– CO
– Nitrates
– Nitrogen oxide
– Sulfur dioxide
– Sulfur dioxide (trans)
– Suspended particulates
37000-57000
16000
25-41000
25-29000
10000
12-13000
12-13000
Water pollution
• Differentiated EKC evidence by pollutants
• From 3400$ for nitrates to 17700$ for lead
• All other pollutants (cadmium, fecal
coliforms, oxygen, arsenic, sulfur
dioxide…) around 8000-13000
Deforestation
• Some studies, but:
– Low data quality
– Deforestation is a complex issue: many drivers play a role
• Evidence for tropical countries, latin america, africa
– 9000 and 8000$ for Latin america and africa in 2001..higher than
most country level around 5000
– Other study 10000-12000 in 2003$
– Other evidence (Panayotou and others) find lower turning points
occuring at 1000-3000$...but we recall that the environmental
peak matters…
– Panayotou also argues that the most dramatic impact overall
occurs between 1000 and 3000 per capita (China, India,
Africa...)
• The first methodological problem for the applied analysis
is how specifying the EKC functional relationship. There
is no consensus on it.
• Some authors use second order polynomial, others have
estimated third and even forth order polynomials,
comparing different specifications for relative robustness.
It is worth noting that neither the quadratic nor cubic
function can be considered a full realistic representation
of the income-environment relationship.
• The cubic implies that environmental degradation will
tend to plus or minus infinity as income increases, the
quadratic implies that environmental degradation could
eventually tend to zero. The issue is thus highly
unresolved.
• Shobee (2004) suggests a third order polynomial
specification as more realistic relationship between
environmental degradation and income per capita. This
supports the credence of a logistic shape, wherein
environmental degradation first accelerates, then
decelerates, and finally falls.
• Marginal environmental degradation is thus not modelled
as constant. The issue still remains highly unresolved,
with the EKC hypothesis relying mainly on empirical
evidence. The theoretical foundations of the EKC are still
ambiguous, though some contributions have emerged
(Andreoni and Levinson, 2001).
• Third or forth level polynomial could also lead to N rather
than U shaped curves, opening new problematic issues
in understanding the income-environment phenomenon
for policymaking.
• This N shape is justified by a non-linear effect by the
scale of economic activity on the environment, which is
difficulty to prove.
• Finally, the use of the income factor only, without
quadratic and cubic terms, would collapse the EKC
analysis to the basic decoupling analysis.
• For a simple presentation of EKC with a discussion of
the core hypothesis see De Bruyn et al. (1998) and Stern
(2004).
• The aforementioned delinking hypothesis is tested by
specifying a proper reduced form usual in the EKC field
(Stern, 2004).
– panel data framework, where the relative fit of fixed effect and
random effect models is compared by the Hausman statistic
– it allows the treatment of unobserved effects..but rare..
– Cross section frame (usually OCSE, world bank
data..)..endogenity issue…causality relationship…???
– The need to move the analysis from cross country to single
country and regional analysis..cross country analysis may hide
some EKC specificity..we just estimate mean values…no policy
value..
The hypotheses on Delinking are thus
tested by estimating a reduced form
regression model (i.e. waste):
• log(waste/N)= 0i + t +
1Log(Consumption/GDP) it + 2
Log(Consumption/GDP)2 it + 3
Log(Consumption/GDP)3 it + ei
• Where the first two terms are intercept
parameters, which vary across countries
and years (in a panel frame).
Literature review
• An extensive literature has developed since the early 90s…
• Biased on emissions at macro level…good availability…but rough
data..need of panel regional studies…
• Let’s see and comment some key works..
• The main objective is the finding of a turning point…
• …but even the presence of a demonstrated turning point may not
assure a Sustainable path..
– Irreversible losses over a certain environmental impact (critical
threshold)
– Developed and non developed countries…may the world economy
sustain the impact of a EKC path in developing countries equal to that
experienced by western economies?
– The problem of consequential developemnt….with scarce resources…
Innovation must be transfered in order to “cut” the developing countries
EKC…this is a crucial issue within Kyoto
List and Gallet (1999), Ecolec
• Sulfur dioxide and nitrogen oxide for 1929-1994,
panel data
• US State level
• They underline:
– Evidence at US state level
– The issue of “omitted variables”..when including
additional explanatory variables EKC evidence
changes…
– State level evidence is needed..they find that states
with urban areas and more densely populated show a
turning point at lower income levels..more policy
attention?
Harbaugh et al (2001), NBER
paper
• They also test whether EKC evidence changes when the
set of explanatory factors is extended, in addition to
income terms…
–
–
–
–
Policy
Country/state features (weather, pop density, industrial/rural..)
Other control variables
TRADE intensity
• Evidence may change even when
– the set of countries is changed
– The number of years is different
• They show that for SO2, smoke and total susp part. The
GDP/pollution relationship is sensitive to both sample
seelction and empirical specification
• No EKC evidence
TRADE
• Some authors claim that for some
pollutants the EKC evidence may only
imply that more polluting activities are
moved to other countries  from western
to developing countries
Markandya et al (2004), FEEM
• This is a EU12 specific study using a long time
series..probably the most effective way of estimating
EKC
• Sulfur emissions 1850-1999: top value for 9 countries
during 1970-1985
• Panel analysis:
– Best fir with fourth order polynomial
– Two turning points: 7000$ and 25000$ (lower emissions) per
capita income
– Individual country analysis show different model fits and different
turning points
• Then, UK only to account for the effect of
Acts/regulations on emissions..including a dummy in
policy years..no impact
McPherson et al (2005), ECOLECON
•
•
•
•
EKC and Threatened species..new issue
Cross country data..poor data availability
113 countries in 2000
Dep var: % of threatened species in 2000 + extinct in
1990-1999
• Results:
– For birds and mammals, the curve seems rising up to around
10000-15000$, then species classified as threatened decreases
– Population density negatively impacts
– Islamic and post-communist and communist countries have
more species as threatened..property rights definition? Green
policies in democratic countries…this opens a new public choice
political issue (growth-democracy-environmental quality..what is
the causal link?)
European waste case study
(Mazzanti and Zoboli, 2005)
• Results indicate that stronger efforts are needed in order
to increase the waste-oriented efficiency of economic
processes. They confirm the fact that, concerning
primary waste issues, even relative de-linking is
questioned on a average European level.
• Environmental policy makers should remain hesitant
about the idea of economic growth as self-sufficient
driving force of environmental improvements, the reality
is much more complex for most environmental
indicators.
• The income elasticity of primary non hazardous waste
flows in observed European countries is likely to be, or
become in the medium term, less than one, but it is
certainly not negative
Final points
• Different polynomial specifications are tested by
including as
• (i) dependant variable waste per capita and waste in
absolute terms,
• (ii) independent variables either household consumption
or GDP per capita, thus testing the hypothesis which
indicates that consumption is a more appropriate driver
for waste.
• In fact, recent studies (Rothman, 1998; EEA, 2003a,b)
point out that for municipal and packaging waste the
proper economic driver/indicator is not GDP, but
household consumption instead. This is a key issue on
both conceptual and statistical grounds.
..from EKC to environmental
Innovation
• The EKC framework has confirmed the need of
triggering innovation and then transferring innovation to
industrializing countries…
• since property rights definition, market functioning,env.
policy and income are four main sources of innovation
• Interesting study by Komen et al (1997) on the
relationship between income and public R&D in
environmental protection activities (not abatement)
– They find an income elasticity not different from one
– It shows the relevancy of the income  R&D  innovation 
growth + SD complex set of dynamic links
• You may find papers at www.deit.economic.unife.it 
publications
• Two 2005 papers by Mazzanti-Zoboli, also some 2004
papers on other issues…
• You may also look at my course page in
www.economia.unife.it to download publications
• My email is [email protected]
• Good sites:
–
–
–
–
www.feem.it
www.rff.org
CSERGE/UCL site
Good journal is Ecological economics, JEEM, Env&ResEcon,
Land economics