Transcript Rome pension funds

The Effects of Fiscal Policy on the Environment
Thesis Defence
by
Epameinondas Paizanos
Department of Economics, University of Thessaly
Supervisor
Professor George E. Halkos
Volos, 30 October 2015
This research has been co-financed by the European Union (European Social Fund – ESF) and Greek national funds
through the Operational Program "Education and Lifelong Learning" of the National Strategic Reference Framework (NSRF)
- Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.
1
Introduction
• Despite the immense importance of government
expenditure and government revenues in many
economies worldwide, little effort has been so far
devoted to the study of the impact of fiscal policy
on environmental quality.
• Focus on the effect of government expenditure,
but also considers impact of government revenues.
2
Contribution
Objective is to examine dimensions of this relationship
that have not been sufficiently considered in previous
work.
1. Few studies, but growing body of literature.
2. Estimate both the direct and indirect effects of
government
expenditure
on
environmental
degradation and thus, report the total effect.
3. Examine how the magnitude of the direct effect of
government expenditure on environmental quality
may differ conditional on the level of economic
development and the quality of institutions in a
country.
3
Contribution (cont’d.)
4. Provide estimates of the effect of fiscal policy on
different indicators of environmental pollution,
which can be directly compared.
5. Offer
insights
concerning
the
short-term
interrelationships
between
government
expenditure and government revenues with air
pollution.
6. The importance of employing appropriate dynamic
econometric techniques in this framework is
emphasized and the dynamic nature of the
examined relationships is explicitly taken into
consideration throughout the empirical analysis.
4
Fiscal Policy - Environment
• Mechanism and effect may differ according to whether
pollution is production- or consumption-generated
(McAusland, 2008).
Production-generated pollution (Lopez et al., 2011)
• Scale effect (change in environmental pressures due
to the effect of government expenditure on economic
growth),
• Composition effect (increased human capital
activities instead of physical capital intensive industries
that harm environment more),
• Technique effect (due to higher labor efficiency)
• Income effect (increased income raises demand for
improved environmental quality).
5
Fiscal Policy – Environment (cont’d.)
Consumption-generated pollution (Galinato and Islam, 2014)
• Scale (Income) effect (spending on health and
education increases the current and future income of
households and leads to an increase of consumption
pollution)
• Regulation
effect
(government
expenditure
encourages the development of institutions and
therefore the establishment and enforcement of
environmental
regulations
which
enhance
environmental quality) - Fullerton and Kim, 2008.
6
Fiscal Policy – Environment (cont’d.)
Other related channels
• Governance quality (enhanced institutional quality
reinforces the alleviating effect of government
expenditure on pollution) – Galinato and Islam 2014; Bernauer and
Koubi, 2009.
• Special interest groups (influence the strictness of
environmental regulation the government imposes) Mueller and Murrell 1986; Bernauer and Koubi, 2013; Galinato and Islam,
2014; Sobel, 2001.
7
Fiscal Policy – Environment (cont’d.)
Empirical evidence
• Ambiguous effect on environmental degradation.
• Positive effect (Koubi Bernauer, 2013)
• Mixed evidence (Frederik and Lundstrom, 2001)
• Negative effect
(Lopez et al., 2011; Lopez and Palacios, 2014;
Galinato and Islam, 2014)
8
Estimating the total effect of government spending
• Government size has been found to influence
prosperity (Ghali, 1998; Bajo-Rubio, 2000; Folster and Henrekson,
2001; Bergh and Karlsson, 2010)
• This in turn may impose an indirect effect on
pollution, depending on the shape of the
Environmental Kuznets Curve (Grossman and
Krueger, 1995).
• This study is the first to distinguish and empirically
estimate both direct and indirect effects of fiscal
spending on the environment.
Prof. George E. Halkos & Epameinondas A. Paizanos
9
Data
• Panel data for 77 countries with full set of SO2,
CO2, share of Govshare, GDP/c and other
explanatory variables for 1980-2000. 1617
observations per variable.
• SO2 emissions production-generated with local
impact.
• CO2 emissions mix with global impact.
10
Methodology
Model
• Control variables in Eq. (1) include GDP/c (scale
effect), household final consumption (income
effect) and democracy level (proxy for
environmental regulation).
• Thus, coefficient of GOVSHARE (i.e. direct effect)
mainly captures composition effect and part of
technique effect.
• Equation 2 is augmented Solow model.
11
Estimating the total effect of government spending
• The total effect of government spending on pollution
can be expressed as:
12
Methodology
• FE are used instead of RE (Hausman test).
• Since our panel consists of large N and T
dimensions, dynamics and non-stationarity are
taken into account by employing the Dynamic
Fixed Effects (DFE) estimator proposed by
Pesaran and Smith.
• To address bias occurring from potential
endogeneity between GOVSHARE with pollution
and income respectively, we apply A-B GMM.
13
14
Fisher-type Philips-Peron tests allowing heterogeneity of autoregressive
parameters
In all cases variables are I(1)
15
• Application of DFE requires variables cointegrated (LR relationship).
DFE is applicable.
• We reject Ho: no-cointegration in 5/8 cases for SO2 and in 6/8 in CO2.
Evidence of cointegration.
16
17
18
• The negative sign of indirect effect occurs from the
positive relationship between income and pollution
at the median income level (upward section of
EKC).
19
• The total effect of government share on SO2/c is negative
for low levels of per capita income and then turns to
positive (above $10,809), while the total effect on CO2/c is
also negative but becomes positive only for very high
income levels (above $16,438).
20
Sensitivity analysis
Relative tests indicate that the results are robust to :
• Inclusion of a GOVEXP composition variable.
• Omission of time-variant variables.
• Use of different model specifications.
• Inclusion/exclusion of extreme observations.
21
Magnitude of the direct effect
• Environmental quality is influenced by various other
factors, apart from fiscal policy and economic growth,
including political institutions, population, trade and capital
stock (Grossman and Krueger, 1995; Halkos, 2013a; Bernauer and
Koubi, 2009; Zhu and Peng, 2012; Cole and Elliott, 2003).
• Some of these characteristics may interact with
government expenditure and influence its effect on
environmental quality.
• It is highly unlikely that the direct effect of government
spending on pollution is independent from country specific
characteristics.
22
Magnitude of the direct effect (cont’d.)
Institutional quality
• Small elites in autocratic regimes oppose the
establishment of environmental regulations (Bernauer
and Koubi, 2009; Lake and Baum, 2001; Niskanen, 1997; Farzin and
Bond, 2006).
• Corruption
reduces
the
stringency
of
environmental regulations (Fredriksson and Svensson, 2003;
Fredriksson et al., 2003; Cole, 2007; Halkos and Tzeremes, 2013).
• The median voter, in democracies, faces a
relatively lower burden from environmental policies.
23
Magnitude of the direct effect (cont’d.)
Level of economic development
In developed countries:
• the income effect exceeds the scale effect (Islam
and Lopez, 2015).
• more demand for environmental quality since the
demand for other public goods has already been
satisfied (Frederik and Lundström, 2001;
Martinez-Mongay, 2002).
• government expenditure more likely to be
targeted at education and R&D spending.
24
Data & Methodology
• Panel data for 94 countries. SO2, N2O, CO2, NOx, Govshare,
GDP/c and other explanatory variables for 1970-2008. 3.525
observations per variable.
• Use of dynamic estimation methods (slow adjustment) –
Perman and Stern, 1999
• Partial Adjustment Model.
• Dynamic Fixed Effects.
25
Partial effects
26
Magnitude of the indirect effect
27
Magnitude of the direct effect (cont’d.)
28
Magnitude of the direct effect (cont’d.)
29
Magnitude of the direct effect (cont’d.)
30
Magnitude of the direct effect (cont’d.)
31
Short-run effects
• The empirical studies use reduced-form models and
estimate the long-term effect of government spending.
• This section draws structural conclusions using Vector
Autoregression methods based on minimal hypotheses
about the signs of the impacts of certain shocks (Mountford
and Uhlig, 2009).
• Examines for the first time how different realistic policy
implementation scenarios, that aim to stimulate the
economy, may also impact environmental quality.
• Quarterly data for 10 macroeconomic and 2
environmental variables for the U.S. economy, for the
period 1973-2013. 164 observations per variable.
32
Short-run effects
• Following Mountford and Uhlig (2009) the computations
are performed using a Bayesian approach.
• The sign identification procedure starts with a reduced
form VAR as:
• The VAR system has two lags (based in AIC and SC
criteria).
• No constant term or a time trend included.
• Logarithm for all variables is used, except for interest
rate.
33
Short-run effects
• The business cycle and monetary policy shocks are identified first.
• The basic fiscal policy shocks are then identified by requiring that
they are orthogonal to both business cycle and monetary policy
shocks.
• No sign restrictions imposed on the reaction of the environmental
and macroeconomic variables of interest to fiscal policy shocks.
34
Short-run effects
• This identification strategy, combined with the
measure of government expenditure used, serves
the separation of changes in the variables of interest
that are the result of actual fiscal policy shocks,
rather than capturing their variation in response to
the business cycle shocks.
• To enhance the identifying power of the model, the
responses of the fiscal variables are restricted for
four quarters after the initial shock
35
Short-run effects
36
Short-run effects
37
Short-run effects
38
Short-run effects
39
Conclusions
• Government expenditure has a negative or nonpositive direct effect on the different indicators of
environmental degradation.
• The alleviating direct effect of government spending on
pollution is significantly reinforced in developed
countries and democratic regimes.
• The direct effect is greater in significance and in
magnitude on pollutants that are characterized by shorter
atmospheric life times, local geographical range and
therefore more immediate impact on human health.
• The estimated direct effect is larger in significance and
magnitude
on
production-generated
pollution
compared to consumption-related pollution.
40
Conclusions (cont’d.)
• The alleviating direct effect of government expenditure
on environmental degradation can be considerably
enhanced by targeting spending on specific functional
categories, such as spending on public goods and
environmental protection.
• The total effect of government expenditure on
environmental quality depends on the government
spending-growth
and
growth-environment
relationships (i.e. the indirect effect).
• An attempt to stimulate the economy through tax-cuts is
associated with an increase in consumption-related CO2
emissions in the short-run, while there is little evidence of
an effect on production-generated CO2 emissions.
41
Policy implications
• Fiscal spending may be used to complement the efforts to
improve environmental quality, rendering these efforts easier
and more cost efficient.
• The weak direct effect of government spending on
environmental quality in developing countries and autocratic
regimes can be enhanced by enforcing property rights on
natural resources which may promote the internalization of
environmental externalities.
• There is a need for technological and knowledge diffusion
to developing countries in order to advance the abatement
methods used and encourage the use of environmental
cleaner production methods.
• The mitigation of pollutants with more global impact, requires
the adoption of international environmental treaties which
will internalize such externalities.
42
Policy implications (cont’d.)
• Provides reassurance to macroeconomic policy makers
that increasing government expenditure does not deteriorate
CO2 related air pollution and in fact is associated with
improvements of environmental quality.
• Governments in developed countries should restructure
government expenditure composition and reform tax systems
in order to ensure that these are not linked with a
slowdown of economic growth. This can be achieved by
promoting spending in categories that alleviate market failures
such as investments in human capital, R&D and health
(Lopez et al., 2011; Fakin, 1995; Lee, 1996).
• A fiscal expansion based on tax-cuts increases consumptiongenerated CO2 emissions in the short-run and should be
accompanied by stricter environmental regulations.
43
Limitations and future research directions
• Lack of data on the composition of government
expenditure for a large sample of countries and for a
long period of time.
• Analysis of the short-run link between fiscal policy
and environmental degradation is also limited by
data unavailability in most countries and thus
prevents a more representative study on this
relationship.
44
Limitations and future research directions (cont’d.)
• The analysis should be conducted in a joint
framework with two other bodies of literature.
• Lack of theoretical models that examine the
relationship between fiscal policy, output and
environmental quality.
• For the establishment of such models, the empirical
finds occurring from this thesis can provide a useful
starting point.
45
• Thank you for your attention!
46