Is democracy enough to get early turn of the environmental Kuznets curve in ASEAN countries?

Abstract

This research article investigates the impact of democracy on peak turning point of the environmental Kuznets curve in the context of the Association of South East Asian Nations (ASEAN) countries that have different levels of development and democracy. For this purpose, time-series data for the period of 1980–2014 have been employed. The empirical findings support the existence of the environmental Kuznets curve for CO₂ emission in the ASEAN countries. It is also found that democracy measured by civil liberties delays the turning point of the environmental Kuznets curve and, therefore, increases the environmental cost of economic growth. The research output provides meaningful inputs to formulate an effective public policy for the protection of the environment in the ASEAN countries.

Keywords

Environmental Kuznets curve pollution democracy

Introduction

The relationship between income and environmental degradation was empirically examined during the start of the 90s. The output of these studies formulated the so-called environmental Kuznets curve (EKC) hypothesis. This hypothesis postulates that economic growth eventually overcomes the environmental damages of early stages of economic development.^1–4

EKC states a nonlinear relationship between economic growth and pollution as shown in Figure 1. During the early stages of the economic development, economic growth deteriorates environmental quality; however, as growth process continues, it generates conducive conditions for a better environment. Economic growth, therefore, can be considered a cause as well as remedy to the environmental problems.

Figure 1.

EKC relationship between economic development and pollution.

The rationality of the EKC relationship can be assessed on both supply and demand sides. The supply side effects are disintegrated into scale, composition, and technique effects. The scale effect starts with the surge of economic activities by utilizing more natural resources that results in more pollution, while the composition effect refers to the transition in production structure of an economy from agrarian to industrial and then to services sectors. In addition, composition effect reallocates the resources along with upsurge in environmental degradation in the first stage and improvement in environment in the second stage. Finally, the technique effect ensures better methods of production due to technological advancement. The outdated machineries are substituted with updated and cleaner technologies that lead to a decline in the level of pollution.

Whereas on demand side, people are worried about the improvement of material welfare at the early phase of economic development. They are less concerned about the quality of the environment. However, as their income increases, their inclination to pay for a cleaner environment improves, and the institutions that regulate environment also develop more effective approaches for regulating the pollution.⁵ These supply and demand side changes lead to an inverted U-shaped relationship between economic growth and pollution.

EKC has been empirically tested and critically reviewed in all parts of the world. Though some of the researchers agree about its existence, they have disagreement on the shape and turning point of the EKC. From the theoretical and empirical reviews on the EKC, Copeland and Taylor⁶ concluded that the economic development and environment lack any predictable and simple relation.

The EKC, as an optimal growth path and high income as a prerequisite for environmental improvement, has fascinated the development and environmental policies of the world.⁷

The EKC notion also signifies tradeoff between economic growth and environment at early stages of economic development. For developing countries, this tradeoff has important implications. Economic growth is central to counter unemployment, poverty and hunger; therefore, these countries cannot compromise over economic growth. Thus, the real issue for the developing countries is that how they can maintain economic growth yet minimizing its environmental cost.

The EKC provides an important policy guideline in this regard. The steeper EKC implies high environmental cost of economic growth while the flatter EKC indicates least environmental losses of economic growth. Several factors such as international trade, foreign direct investment (FDI), technology, environmental governance, environmental awareness, and democracy contribute to determine the slope and altitude of the EKC.

The current study, therefore, aims to examine the role of democracy in shaping income environment relation in the Association of South East Asian Nations (ASEAN) countries. It is the least searched area in the EKC literature especially in the context of ASEAN countries as they have different levels of development and democracy.

According to Deacon and Mueller,⁸ the political system of a country, either democratic, semi-democratic, or autocratic, plays a critical role in the allocation of environmental resources. When democratic governance and rule of law are not well established, prospective corruption may boost rent-seeking behaviour among government officials and businessmen. Similarly, when government tends to represent the interest of selected groups rather than of entire population, the provision of a public good like environment may be compromised. Those who defend the role of liberal democracies for environment claim that there is no other system than democracy that guarantees all basic human rights to individuals including the right of a healthy environment.

On the other hand, the critics of liberal democracy hold that democracy provides the individuals with opportunities to escape their environment responsibilities if they choose to do so. The young democracies also lack necessary institutional arrangements that can foster peoples' participation and accountability. The nascent institutions, therefore, fall prey to the preference of the elites rather than to protect the interests of broad-based population. Chang and Cho^9,10 and Midlarsky¹¹ held the view that democracies in developing countries did not generate supporting conditions for a cleaner environment. Hence, there is no conclusive argument about the role of democracy in shaping the income environment relation. Some take it as a prerequisite of environmental performance while others consider it as a plague for environmental quality.

The ASEAN countries have several environmental problems. According to literature,¹² increased industrialization and urbanization in the 1990s and 2000s generated severe environmental problems like air and water pollution and accumulation of urban wastes in the ASEAN countries. The Global Climate Risk Index (2016) also indicated that ASEAN countries are the most vulnerable to environmental changes. Most of these countries are island countries and are exposed to the risk of rising of the ocean level. Moreover, World Bank (2015) observed momentous increase in CO₂ emission over the last three decades in the ASEAN region. The countries like Singapore, Malaysia, Indonesia, Thailand, and Vietnam have a significant increasing trend of carbon emission.

While democracy made inroads in the ASEAN region after colonial rule ended, it has been struggling to take root as authoritarianism has left deep legacy. After the human right declaration adopted unanimously in 2012, the ASEAN countries sought to believe in democracy as the engine for economic growth. Though socialist states like Laos and Vietnam have embraced capitalism on economic front, they are still ruled by communist parties and have not abandoned the authoritarianism. Similarly, Malaysia and Singapore, despite economic modernization, have kept democratizing at bay. Likewise, the other ASEAN states in the past have experienced democratization such as Philippines in 1986, Cambodia and Thailand in the early 1990s, and Indonesia in 1998; however, none of them have fully entrenched the liberal and democratic values.

Given this performance of the ASEAN countries on democracy and environment, the current study aims at examining the role of the former to decrease the environmental losses of economic growth. In other words, the study tests the hypothesis: Has democracy potential to get an early turn in the EKC in ASEAN countries?

Literature review

Various empirical studies provide contrasting results on the relationship between the level of democracy and environmental degradation. For instance, Congleton¹³ found a positive impact of liberal democracies on the environment from the analysis of democratic and non-democratic countries of the world. Torras and Boyce¹⁴ pointed out that more equitable distribution of political power resulted in better environmental quality. Barrett and Graddy¹⁵ also found that political liberties and civil rights have a positive impact on the environmental quality of a country. Similarly, Neumayer¹⁶ verified a positive relationship between democracy and environmental commitment and recommended that more democracy around the world would result in more commitment on the environmental front.

While Li and Reuveny¹⁷ found a non-monotonic impact of democracy on the environment that varied across different degrees of the democracy, they, therefore, concluded that the impact of democracy on income and environment relation was dependent on how much democracy was matured in that country. They used CO₂, land degradation, deforestation, and organic pollution in water as a measure of environmental quality and regressed them against the level of democracy and Gross Domestic Production (GDP) per capita in 143 countries for the period of 1961–1997.

Aubourg et al.¹⁸ pointed out that the level of democratization was so important that it could change the turning point of the EKC. Likewise, Ward¹⁹ indicated that stable democracies performed better on sustainability indicators as compared to stable autocracies. Similarly, Bernauer and Koubi²⁰ provided support to the stance that the degree of democracy has an independent positive impact on air quality in a country. Moreover, Bernauer and Koubi²⁰ highlighted that more liberal political system is conducive to environmental quality.

Buitenzorgy and Mol²¹ again confirmed that democracy had larger explanatory power than income to explain the variation in deforestation rate. Arvin and Lew²² found the same evidence that democracy was beneficial for the improvement of environment; however, this outcome varied across various pollutants.

Vanek²³ analyzed the impact of the operations of capitalist and democratic firms on the environment. He found that the democratic firms were more concerned about the natural environment as compared to the capitalist firms. Moreover, he also indicated that capitalist firms tend to avoid environmental-related costs and, therefore, destroy the environment. Farzanegan and Markwardt²⁴ revealed that democratic development in the Middle East and North African (MENA) countries was beneficial for the curtailment of environmental problems. Similarly, Dipietro²⁵ showed a positive impact of democracy on total national environment performance.

Equally, Fredriksson and Neumayer²⁶ argued that democratic capital (i.e. stock of historical experience with democracy) has more sound impact on the environment rather than the current level of democracy. They claimed that higher level of democratic capital was highly correlated with the stringency of environmental regulations. Therefore, as democracy gets consolidated in developing countries, it would exert a positive impact on the environment. Furthermore, Sjöstedt and Jagers²⁷ found that more democratic countries were more successful in protecting the marine life.

You et al.²⁸ examined the role of better democracy to reduce emissions among the nations with most and least carbon emission. Their results showed that the effect of democracy on CO₂ emissions was not homogeneous, rather more democracy appeared to reduce emissions in most emissions nations. From a panel of 47 transition economiesPolicardo²⁹ also concluded the same that democracy is conducive to control environmental degradation in the process of economic development. He indicated that democracy had better environmental targets than dictatorship.

Buitenzorgy³⁰ empirically tested the political EKC in Indonesia. Political (EKC) implies an inverted U-shape relation between environmental quality and level of democracy. He confirmed that pollutants like CO₂ and biochemical oxygen demand follow an EKC path with the development of democracy. He also supported that improved democracy may play more significant role than increased income to reduce the environmental degradation. Sulemana et al.³¹ also came up with the similar findings that democracy significantly affected the Particulate Matter (PM₁₀) emissions in the African countries. Further, Ghodrati et al.³² also revealed that improvement in democracy in the countries that have low and medium human development indices can lead to improvement in environmental quality.

On the contrary, Chang and Cho⁹ claimed an inverse relationship between democracy and environment in 27 Asian countries for the period of 1991–2001. They found that democracy measured by civil liberties had increased the CO₂ emission in these countries. Similarly, López and Galinato³³ suggested that countries at early stages of democratic process had unstable governance institutions that created negative implications for the environment.

Scruggs,³⁴ after going through the 58 studies that directly deal with the influence of democracy on environmental performance of a country, concluded that economic growth and wealth have more sound impact on environmental performance of a country as compared to the impact of democracies. Arvin and Lew²² did not find any uniform relation between the state of democracy and state of the environment. Similarly, Faure et al.³⁵ raised the doubts over the role of democracy for the improvement of environment. They argued that, at the early phases of economic development, citizens do not demand investments for environmental protection at the cost of basic public goods like health or education. This lack of pressure from the public on democratic governments results in least investment to improve the environment. Therefore, at the early phases of economic growth, democratic developing countries are not a better performer on environmental front.

Halkos and Tzeremes³⁶ pointed out the same that the regional variation and difference in the economic development of the countries determine the way how different governance indicators affect carbon emission. Likewise, Lv³⁷ pointed out that democracy reduces carbon emission only if the economy has reached a certain level of income. He deduced these results by analyzing the interaction effect of democracy and GDP on carbon emission in 19 emerging economies for the period of 1997–2010.

Vadlamannati and Gaarder³⁸ contributed to the debate by analyzing the contentious relationship between environmental degradation and regime type in 139 countries for the period of 1990–2012. They also found that democracy was linked with lower level of forest coverage. Their analysis further revealed that positive impact of democracy on environment is conditional upon the certain level of economic development. Similarly, Povitkina³⁹ argued that impact of democracy on environment is limited due to widespread corruption in developing countries that reduces the capacity of the democratic governments to reduce CO₂ emission and to meet climate targets. He investigated the environment and democracy linked in 144 countries over the period of 1970–2011.

Hence, empirical literature on the role of democracy in income–environment relation does not have empirical and theoretical consensus. Moreover, scant of the studies have examined the impact of democracy on income–environment relation in the context of ASEAN countries.

Model and data

Model

EKC postulates a nonlinear relationship between income and pollution. The empirical studies on the EKC have used the following model proposed by literature⁴⁰:

Y_{it} = β_{o} + β_{1} X_{it} + β_{2} X^{2} it + μ_{it}

(1)

where Y_it is pollution and X_it is income.

β_{0}

is the constant term, and

β i

are coefficients of independent variables. The µ_it is the error term that captures the variation in Y_it that is not explained by

X_{it}

. While i = 1,2,3.….…. n countries, t = 1,2,3,4…… t years. The EKC relation between income and pollution would exist if β₁ > 0 and β₂ < 0.

As this study aims to examine the role of democracy to determine the income–environment relation in the context of ASEAN countries, therefore, in equation (3), $Dem o_{it}$ represents the democracy variable. Prior literature indicates that pollution is determined by many factors other than income, therefore, to avoid any model miss-specification, two important determinants of pollution such as energy consumption (EC) and FDI are included in equation (2) as control variables.

Y_{it} = β_{0} + β_{1} X_{it} + β_{2} X^{2} it + β_{3} D em o_{it} + β_{4} F D I_{it} + β_{5} E C_{it} + μ_{it}

(2)

FDI is considered as one of the most important determinants of the pollution in the literature of environmental economics. There exist two contrasting views about the impact of FDI on pollution. One is halo effect hypothesis that claims that FDI spurs economic activities, environmental efficiency, and technological improvement, therefore causing a positive environment spill over.⁴¹ The other extreme is pollution haven hypothesis (PHH) that postulates a negative role of FDI for pollution in developing countries. According to the PHH, those developing countries are the main recipient of the FDI that have lax environmental regulations and have worse environment.^42,43 The empirical output of the studies about the impact of FDI on pollution is mixed.^44,45 EC is another important determinant of pollution identified in the literature. There are numerous studies that included EC while testing the link between economic growth and environment like.^46–50

Most of the empirical studies that advocate the role of political and institutional factors in income–environment relation use these factors as dummy variables. In this specification, dummy variables cannot affect the turning point of the EKC, rather they can only change the altitude of the EKC. Resultantly, these studies are unable to address the issue of the different turning points of the EKC, as pointed out by the EKC literature. This study, therefore, includes democracy variable in the EKC model interactively with income so that turning point will become context specific as suggested by literature.⁵¹ This specification provides a way to empirically investigate the different turning points of the EKC due to various level of democracy. The resulting model will be as follows:

Y_{it} = β_{0} + β_{1} X_{it} + β_{2} X^{2} it + β_{3} {(X * demo)}_{it} + β_{4} F D I_{it} + β_{5} E C_{it} + μ_{it}

(3)

where

X * demo

is an interaction term that has the following interpretation. Given the income level, pollution will increase with the increase in the level of democracy, while, given the level of democracy, pollution tends to rise due to increase in income. To determine whether interaction term matters in this model, Wald test for zero restriction of the parameter for interaction term has been used.

In equation (3), democracy affects the environment indirectly by affecting the economic growth. This indirect effect is assumed to influence the turning point of the EKC owing to its effect on GDP. This specification is very important in terms of tracing out the true impact of democracy on the turning point. According to Aubourg et al.,¹⁸ this model allows locating the turning point GDP values inclusive of democracy indicators. With this specification, the formula for determining the GDP per capita at the turning point will be

Turning point income level = \frac{β_{1} + β_{3} {Democracy}_{it}}{2 β_{2}}

(4)

Data

Depending upon the availability of data, six ASEAN countries namely Singapore, Malaysia, Indonesia, Thailand, Vietnam, and Philippines have been selected for investigation over the period of 1980–2014. As per usual practice in the EKC literature, CO₂ emission in “metric ton” has been taken as a measure of environmental degradation and GDP per capita as a measure of income. The data for CO₂ emission have been taken from International Energy Statistics (2016), while per capita GDP data have been obtained from the World Economic Outlook 2016. The data about EC (measured in kg of oil equivalent per capita) and FDI have been collected from World Development Indicator 2015.

For the measurement of democracy, present study uses “Freedom in the World” a yearly survey report prepared by Freedom House 2016. It measures the degree of political rights and civil liberties in all the countries of the world and produces scores for the level of civil liberties and political rights for each nation, and then these scores are combined into an index. These scores are measured on one to seven scales where one represents the highest degree while seven is the lowest. However, to simplify the interpretation of the results, scale has been reversed by subtracting each value from seven. The zero, therefore, represents the lowest level of democracy, and seven represents the high level of democracy. Table 1 provides a descriptive summary of the data used in this study.

Table 1.

Description of the used variables.

Country	Variable	Mean	Standard deviation	Minimum	Maximum
Singapore	CO₂	37.79	13.57	13.47	69.18
	X	24045.54	15654.89	5003.79	56009.54
	EC	4364.49	1384.81	1987.09	7370.40
	FDI	1.77E+10	2.02E+10	1.05E+09	6.85E+10
	DEMO	2.49	0.51	2.00	3.00
Malaysia	CO₂	120.38	72.64	27.97	253.074
	X	4815.89	2852.971	1837.92	11009.10
	EC	1885.34	720.49	872.875	3319.82
	FDI	4.32E+09	3.73E+09	1.15E+08	1.51E+10
	DEMO	2.58	0.50	2.00	3.00
Indonesia	CO₂	275.12	157.81	94.71	640.55
	X	1421.61	1012.18	562.90	3744.53
	EC	641.68	171.16	377.68	866.83
	FDI	4.56E+09	7.66E+09	-4.55E+09	2.63E+10
	DEMO	2.54	1.07	1.00	4.00
Thailand	CO₂	171.60	98.62	37.88	337.26
	X	2564.12	1629.26	718.75	6147.99
	EC	1102.57	504.79	411.70	2087.58
	FDI	4.28E+09	4.20E+09	1.63E+08	1.58E+10
	DEMO	3.31	0.676	2.00	4.00
Philippine	CO₂	59.42	20.36	28.03	98.89
	X	1239.18	633.78	597.84	2844.06
	EC	460.34	24.74	416.70	513.14
	FDI	1.30E+09	1.25E+09	-1.06E+08	5.74E+09
	DEMO	3.80	0.63	2.00	5.00
Vietnam	CO₂	66.22	56.00	16.81	194.45
	X	642.12	524.71	97.16	2048.95
	EC	400.51	149.92	266.33	690.65
	FDI	2.59E+09	3.24E+09	-80000	9.58E+09
	DEMO	0.86	0.88	0.00	2.00

Results and discussion

The standard ordinary least square regression technique with non-stationary time series leads to spurious regression and may cause invalid inference of the empirical results.^52,53 It is therefore, imperative to examine the stationary properties of each variable before estimating the relationship between the variables. For this purpose, panel unit root tests by Levin et al.⁵⁴ (LLC) and (IPS)⁵⁵ have been used. After finding that the variables are co-integrated in the panel, the next step is to estimate long-run cointegration vector. The long-run coefficients are estimated using fully modified least square (FMOLS) estimation method developed by literature.⁵⁶ FMOLS is estimated with non-parametric approach that includes the alterations to tackle the serial correlation and endogeneity problem.

The analysis starts with description of the variables used in the extended EKC model as shown in Table 1. The high values of standard deviation as compared to mean values of the dependent and independent variables indicate substantial variation in the variables.

The precondition using panel cointegration tests is to decide whether the time series variables in the panel have a unit root problem. The panel unit root confirms that all the variables must be either stationary at level I (0) or stationary at first difference I (1) The Im, Pesaran and Shin (IPS) and Levin, Lin and Chu (LLC) unit roots tests for each time series variable are summarized in Table 2. The test statistics of LLC and IPS fail to reject the null hypothesis implying that all the variables in the panel are non-stationary. However, taking the first difference of all the variables, both LLC and IPS reject the null hypothesis and accept the alternate hypothesis that variables are stationary. Hence, it can be concluded that all the variables in the panel are integrated of order I (1).

Table 2.

Results of panel unit root tests.

Variable	LLC		IPS
Variable	Level	First difference	Level	First difference
CO₂	4.72	-4.17	6.78	-6.19
CO₂	[1.00]	[0.00]a	[1.00]	[0.00]a
X	6.61	-4.18	8.15	-4.19
X	[1.00]	[0.00]a	[1.00]	[0.02]a
X ²	9.89	-2.25	9.10	-1.87
X ²	[1.00]	[0.00]a	[1.00]	[0.05]a
FDI	3.16	-9.56	3.26	-12.76
FDI	[0.99]	[0.00]a	[0.99]	[0.00]a
EC	2.42	-5.10	3.16	-6.15
EC	[0.99]	[0.00]a	[0.99]	[0.00]a
Demo^*X	5.01	-6.13	6.28	-5.81
Demo^*X	[1.00]	[0.00]a	[1.00]	[0.00]a

Note: The lag selection for every variable is based on Akaike Info Criterion. LLC and IPS tests for all the series include a constant as an intercept.

Rejection of the null hypotheses of a unit root at the 5% significance level.

With the robust proof that every variable is non-stationary and become stationary at first difference, the panel cointegration tests have been applied to examine the long-run equilibrium relationship among the variables. The results of various cointegration tests suggested by Pedroni⁵⁷ and Fisher⁵⁵ are presented in Table 3. The five Padroni tests’ statistics are well significant at 5% critical values and, therefore, reject the null hypothesis of no cointegration. The Fisher test also rejects the null hypothesis of none and at most one cointegration vector. Thus, the alternate hypothesis of one and more than one cointegrating vectors is accepted. Resultantly, panel cointegration tests strongly support the existence of long-run equilibrium cointegrating relation among CO₂ emission, per capita GDP, FDI, EC, and democracy in ASEAN countries.

Table 3.

Results of panel cointegration tests for equation (3).

Tests		Statistics	Probability
Panel V statistic		1.92	0.03^a
Panel PP statistic		−2.57	0.01^a
Panel ADF statistic		−3.00	0.00^a
Group PP statistic		−3.38	0.00^a
Group ADF statistic		−4.79	0.00^a
^aRejection of the null hypothesis of no cointegration at the 5% significance level.
Johansen fisher panel cointegration test for equation (3)
Series: CO₂ Y Y² Democracy^aX FDI
Unrestricted cointegration rank test (trace and maximum eigenvalue)
Hypothesized Fisher Stat.^a			Fisher Stat.^a
No. of CE(s)	(from trace test)	Prob.	(from max-eigen test)	Prob.
None	115.6	0.00	82.15	0.00^a
At most 1	47.32	0.00	38.88	0.00^a
At most 2	18.58	0.09	13.38	0.34
At most 3	11.05	0.52	7.76	0.80
At most 4	9.77	0.64	10.06	0.61
At most 5	9.24	0.68	9.24	0.68

Probabilities are computed using asymptotic Chi-square distribution.

After finding the long-run cointegrating relation among the variables, the FMOLS has been applied to estimate the long-run co-integrated relation among the variables. The results are shown in Table 4. According to the FMOLS results, the coefficient on income is positive and statistically significant and the coefficient on squared income is negatively significant. This proves the existence of EKC for this region. The coefficient on two control variables FDI and EC are also found positive and statistically significant. It implies that these variables also contribute to the emission of CO₂ from the ASEAN region.

Table 4.

Results of FMOLS for equation (3).

Variable	Coefficient	Std. error	t-statistic	Prob.
X	0.004383	0.0012	3.80	0.00*
X²	−3.21E−07	−1.29E−08	−24.88	0.00*
Demo*X	0.000505	0.00024	2.08	0.04*
FDI	1.04E−08	2.05E−10	50.65	0.00*
EC	0.021662	0.00289	7.49	0.00*
R-squared	0.768289	S.E. of regression		58.40281
Adjusted R-squared	0.756283	Sum squared resid		658301.4

First-stage residuals use heterogeneous long-run coefficients. Coefficient covariance computed using default method long-run covariance estimates (Bartlett kernel, Newey–West automatic bandwidth, NW automatic lag length). *Indicates the rejection of Null hypothesis that coefficient are not significant at 5% significance level.

The main concern of this research study was to investigate the role of democracy to mitigate the problem of environmental degradation in the context of the ASEAN countries. The study has interesting results in this regard. The efficient on interaction term Demo*X is positive and statistically significant which indicates that given the income level, democracy causes CO₂ emission to rise in this region. The ASEAN countries have a model of controlled democracies where stable and continuous policies have played a significant role to maintain a conducive environment for investment and growth, especially in Singapore and Malaysia. Therefore, given the positive association between income and CO₂ emission, the indirect impact of democracy on CO₂ emission via income is also positive.

The peak turning point income level of the EKC where pollution starts to decline with further economic growth can be calculated by equation (4).

Peak turning point of the EKC= \frac{0.004383 + 0.000505 (Democracy)}{2 (0.000000321)}

The turning point of EKC at zero scores of democracy is observed at $6827.11 per capita. However, considering the average score on democracies of ASEAN countries, the peak turning point of the EKC reaches to $8872.27 per capita. Moreover, at the highest score on democracy, the turning point of the EKC reaches to $123333.33 per capita. These results imply that the environmental cost of economic growth start to increase in the ASEAN countries as they move to a higher score of democracy. Hence, it can be concluded that economic growth with more democracy seems to produce more pollution at the early stages of economic development in this region.

Moreover, most of the ASEAN countries except Malaysia and Singapore are well below the peak turning point income level of the EKC. This implies that the economic growth without any policy measures related to environmental preservation will be accompanied by CO₂ emission in the region.

The Wald test of coefficient restriction has been applied to examine whether the interaction term in equation (3) should be included or not. The results of the test are shown in Table 5. The t-statistics, F-statistic, and Chi-square statistics reject the null hypothesis (H₀) that interaction term does not matter in the equation (3). The alternate hypothesis (H₁), therefore, is accepted that interaction term does matter in the said equation. Hence, Wald test of coefficient restriction supports the inclusion of the interaction term in the EKC model.

Table 5.

Wald tests results.

Test statistic	Value	Df	Probability
t-statistic	2.08	193	0.04*
F-statistic	4.32	(1, 193)	0.04*
Chi-square	4.32	1	0.04*

Null hypothesis: C (3) = 0. The term (Demo*X) does not matter in the model. *Indicates the rejection of Null hypothesis at 5% significance level.

The empirical results of the study support the stance of Faure et al.³⁵ that at the early phases of economic growth citizens do not demand investments for environmental protection at the cost of basic public goods like health or education. This lack of pressure from the public on democratic governments results in least investment to improve quality of environment. Therefore, at the early phases of economic growth, democratic developing countries are not a better performer on environment front. These findings are consistent with the EKC theory that at earlier stages of economic development the focus remains on employment and income growth than environmental protection.

The study also supports the stance of Paehlke⁵⁸ that “the great danger for both democracy and the environment is that while economy and environment are now global in character, democracy functions on only national and local decision levels.” Consequently, the global public good like environment will be ignored at the local level. This is very true in the case of CO₂ emission that has more global repercussions than on local communities. The results are also in line with Scruggs³⁴ who found that “economic wealth and economic growth have the most steady influence on the environmental performance of a country than democracies.” Similarly, the findings are also in line with Arvin and Lew,²² Halkos and Tzeremes,³⁶ and You et al.²⁸ who suggested that democracy was not a uniform solution to environmental problem of the world. Also, the results are consistent with Lv³⁷ who concluded from 19 emerging economies that democracies curtail CO2 emissions only if the countries have reached to a certain per capita income level.

Conclusion and recommendation

The current study contributes to literature by examining the impact of democracy on peak turning point of the EKC in the context of ASEAN countries from 1980–2014. The ASEAN countries have a different level of development and democracy and are also a group of emerging economies. Therefore, studies of the EKC and its connection with democracy may provide important policy implications. The findings support the existence of the EKC for Green House Gases (GHG) emission in the ASEAN countries. It is also concluded that democracy measured by civil liberties can delay the turning point of the EKC.

These results also support the EKC stance that at the early stages of economic development people are more concerned about income and employment. Environment as a normal good is demanded at higher income level. Democracies incorporate these preferences in public policies and, therefore, do not respond to the environmental problems at the early stages of economic development.

While the results are new to the present literature about democracy and environment, yet the current study has its own limitations. For instance, similar to the previous studies on the EKC, underlying study also uses CO₂ emission that is a partial measure of environmental degradation. The future prospective studies may use some comprehensive measures of environment like environmental performance index while examining the democracy environment nexus. Moreover, the results cannot assert any role of democracy for environment unless the results are compared with those from non-democratic countries. Moreover, findings of the study do not imply that autocracy is a remedy to environmental problems of the day. However, we believe in following short run, midterm, and long-run recommendations for the favourable functioning of democracy to protect the environment.

A proactive public policy is needed that can make the flow of the information about environmental problems publicly available and can make people realize that environment is as important as employment and income growth.

The notion of “development versus environment” needs to be transformed to “environment for development.” There is a need to clearly define property rights of natural resources like clean air, water and forests so that free individuals or interest groups may not tend to overexploit them.

The developing countries need democracies that are more responsive to environmental needs of the people and life-threatening problem of environmental degradation. The democracies should have both the citizen watch groups and environmental interest groups that are vital for the favourable functioning of democracies for environment in the developing countries.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Abid Rashid Gillis is an assistant professor in the Department of Economics at the Islamia University of Bahawalpur, Pakistan.

Sallahuddin Hassan is a professor in the School of Economics, Finance and Banking at the UUM College of Business Universiti Utara Malaysia, Malaysia.

K Kuperan Viswanathan is a professor in the Othman Yeop Abdullah Graduate School of Business at the University Utara Malaysia, Malaysia.

References

Grossman

Economic environment and the economic growth. Q J Econ 1995; 110: 353–377.

Panayotou

Empirical tests and policy analysis of environmental degradation at different stages of economic development. Working Paper WP238, Technology and Employment Programme, Geneva: International Labour Organization, 1993.

Shafik

Bandyopadhyay

Economic growth and environmental quality: time-series and cross-country evidence. The World Bank, Washington, DC: World Bank Publications, 1992.

Beckerman

Economic growth and the environment: Whose growth? Whose environment?

World Dev 1992; 20: 481–496.

Selden

Song

Environmental quality and development: is there a Kuznets curve for air pollution emissions?

J Environ Econ Manage 1994; 27: 147–162.

Copeland BR and Taylor MS. Trade, growth, and the environment. Journal of Economic Literature 2004; 42: 7–71.

Chimeli

Braden

JB.

The environmental Kuznets curve and optimal growth. Palisades, NY: Columbia University, 2002.

Deacon R and Bernardo M. “Political Economy and Natural Resource Use.” Departmental Working Papers, Department of Economics, University of California: Santa Barbara, California, 2004.

Chang

Cho

(eds). Democracy and environment: the effect of democratization on environment outcomes in Asia. In: Annual meeting of the midwest political science association, Palmer House Hilton, Chicago, IL, http://www.allacademic.com//meta/p_mla_apa_research_citation/0/8/5/3/7/pages85373/p85373-1.php (2005, accessed 5 January 2010).

10.

Collier

The political economy of state failure. Oxford Rev Econ Policy 2009; 25: 219–240.

11.

Midlarsky

MI.

Democracy and the environment: an empirical assessment. J Peace Res 1998; 35: 341–361.

12.

ADB. Asian Development Bank Sustainability Report 2016: Investing for an Asia and the Pacific Free of Poverty, RPT167995-2, ADB Avenue, Mandaluyong City.

13.

Congleton

RD.

Political institutions and pollution control. Rev Econ Stat 1992; 74: 412–421.

14.

Torras

Boyce

JK.

Income, inequality, and pollution: a reassessment of the environmental Kuznets curve. Ecol Econ 1998; 25: 147–160.

15.

Barrett

Graddy

Freedom, growth, and the environment. Environ Dev Econ 2000; 5: 433–456.

16.

Neumayer

Do democracies exhibit stronger international environmental commitment? A cross-country analysis. J Peace Res 2002; 39: 139–164.

17.

Reuveny

Democracy and environmental degradation. Int Stud Q 2006; 50: 935–956.

18.

Aubourg

Good

Krutilla

Debt, democratization, and development in Latin America: how policy can affect global warming. J Policy Anal Manage 2008; 27: 7–19.

19.

Ward

Liberal democracy and sustainability. Environ Polit 2008; 17: 386–409.

20.

Bernauer

Koubi

Effects of political institutions on air quality. Ecol Econ 2009; 68: 1355–1365.

21.

Buitenzorgy

Mol

AP.

Does democracy lead to a better environment? Deforestation and the democratic transition peak. Environ Resour Econ 2011; 48: 59–70.

22.

Mak Arvin

Lew

Does democracy affect environmental quality in developing countries?.

Appl Econ 2011; 43: 1151–1160.

23.

Vanek

Capitalism, economic democracy, and ecological destruction of our planet. In: Advances in the economic analysis of participatory and labor-managed firms. Emerald Group Publishing Limited, 2011, pp. 289–298.

24.

Farzanegan

Markwardt

Pollution, economic development and democracy: evidence from the MENA countries. Joint Discussion Paper Series in Economics, 2012.

25.

Dipietro

WR.

Does the extent of democracy, the degree of income inequality, or the size of the private sector influence national environmental performance?

Res World Econ 2013; 4: 55.

26.

Fredriksson

Neumayer

Democracy and climate change policies: is history important?.

Ecol Econ 2013; 95: 11–19.

27.

Sjöstedt

Jagers

SC.

Democracy and the environment revisited: the case of African fisheries. Mar Policy 2014; 43: 143–148.

28.

You

W-H

Zhu

H-M

, et al. Democracy, financial openness, and global carbon dioxide emissions: heterogeneity across existing emission levels. World Dev 2015; 66: 189–207.

29.

Policardo

Is democracy good for the environment? Quasi-experimental evidence from regime transitions. Environ Resour Econ 2016; 64: 275–300.

30.

Buitenzorgy

Three essays on democracy, economic development and the environment. 2016. Faculty of Agriculture and Environment, University of Sydney.

31.

Sulemana

James

Rikoon

JS.

Environmental Kuznets curves for air pollution in African and developed countries: exploring turning point incomes and the role of democracy. J Environ Econ Policy 2017; 6: 134–152.

32.

Ghodrati

Harati

Nazari

Democracy and environment quality in selected countries: an application of panel data. Iran Econ Rev 2018; 22: 21–49.

33.

López

Galinato

GI.

Deforestation and forest-induced carbon dioxide emissions in tropical countries: how do governance and trade openness affect the forest-income relationship?.

J Environ Dev 2005; 14: 73–100.

34.

Scruggs

(ed). Democracy and environmental protection: an empirical analysis. In: Annual meeting of the midwest political science association 67th annual national conference, Palmer House Hilton, Chicago, IL, 2009.

35.

Faure

Martin

Zhiping

, et al. Instruments for environmental governance: what works?. Environ Governance Sustainability 2012.

36.

Halkos

Tzeremes

NG.

Carbon dioxide emissions and governance: a nonparametric analysis for the G-20. Energy Econ 2013; 40: 110–118.

37.

The effect of democracy on CO₂ emissions in emerging countries: does the level of income matter?

Renewable Sustainable Energy Rev 2017; 72: 900–906.

38.

Vadlamannati

Gaarder

AR.

Does democracy guarantee (de)forestation? An empirical analysis. Int Area Stud Rev 2017; 20: 97–121.

39.

Povitkina

The limits of democracy in tackling climate change. Environ Polit 2018; 27: 411–432.

40.

Dinda

Environmental Kuznets curve hypothesis: a survey. Ecol Econ 2004; 49: 431–455.

41.

Albornoz

Cole

Elliott

, et al. In search of environmental spillovers. World Econ 2009; 32: 136–163.

42.

Cole

Elliott

RJ.

FDI and the capital intensity of “dirty” sectors: a missing piece of the pollution haven puzzle. Rev Dev Econ 2005; 9: 530–548.

43.

Cole

Elliott

Fredriksson

PG.

Endogenous pollution havens: does FDI influence environmental regulations?.

Scand J Econ 2006; 108: 157–178.

44.

Zarsky

Havens, halos and spaghetti: untangling the evidence about foreign direct investment and the environment. Foreign Direct Investment Environ 1999; 13: 47–74.

45.

Xing

Kolstad

CD.

Do lax environmental regulations attract foreign investment?

Environ Resour Econ 2002; 21: 1–22.

46.

Ang

JB.

CO₂ emissions, energy consumption, and output in France. Energy Policy 2007; 35: 4772–4778.

47.

Richmond

Kaufmann

RK.

Is there a turning point in the relationship between income and energy use and/or carbon emissions?.

Ecol Econ 2006; 56: 176–189.

48.

Apergis

Payne

JE.

CO₂ emissions, energy usage, and output in Central America. Energy Policy 2009; 37: 3282–3286.

49.

Kivyiro

Arminen

Carbon dioxide emissions, energy consumption, economic growth, and foreign direct investment: causality analysis for Sub-Saharan Africa. Energy 2014; 74: 595–606.

50.

Halicioglu

An econometric study of CO₂ emissions, energy consumption, income and foreign trade in Turkey. Energy Policy 2009; 37: 1156–1164.

51.

Webber

Allen

DO.

Environmental Kuznets curves: mess or meaning?

Int J Sustainable Dev World Ecol 2010; 17: 198–207.

52.

Granger

Newbold

Spurious regressions in econometrics. J Econom 1974; 2: 111–120.

53.

Kao

Spurious regression and residual-based tests for cointegration in panel data. J Econom 1999; 90: 1–44.

54.

Levin

Lin

C-F

Chu

C-S.

Unit root tests in panel data: asymptotic and finite-sample properties. J Econom 2002; 108: 1–24.

55.

Im KS, Pesaran MH, Shin Y. Testing for unit roots in heterogeneous panels. Journal of Econometrics 2003; 115: 53–74.

56.

Pedroni

Purchasing power parity tests in cointegrated panels. Rev Econ Stat 2001; 83: 727–731.

57.

Pedroni

Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxf Bull Econ Stat 1999; 61: 653–670.

58.

Paehlke

RC.

Environmental challenges to democratic practice. In: Lafferty WM and Meadowcroft J (eds) Democracy and the environment: problems and prospects. Cheltenham, UK: Edward Elgar, 1996, pp.18–38.