Dynamic common correlated effects of trade openness, FDI, and institutional performance on environmental quality: evidence from OIC countries

Abstract

The study aims to address the dynamic common correlated effects of trade openness, FDI, and institutional performance on environmental quality in OIC countries. Mostly, pollutants like CO₂ and SO₂ emissions are considered as the environmental indicators. However, for this study, we have selected ecological footprint as the indicator of environmental quality. The new econometric approach Dynamic Common Correlated Effects (DCCE) by Chudik and Pesaran (2015) has been used to measure the cross-sectional dependence among cross-sectional units. Results confirm that previous techniques for long panel data, like MG and PMG, give ambiguous outcomes in the presence of cross-sectional dependence. According to DCCE estimation, trade openness, FDI, and urbanization have a positive and significant relationship with ecological footprint while a significant and negative association is found between institutional performance and ecological footprint. The OIC countries must encourage green technology, clean production, and improved institutions for sustainable development and better environmental quality.

Introduction

During the growth process, the ecological quality of any nation degrades swiftly due to decrease in forest cover, deterioration of ambient air quality, and soil erosion (Destek and Sinha 2020). Mostly, pollutants like carbon dioxide (CO₂) and sulfur dioxide (SO₂) emissions are considered as the ecological or environmental indicators. However, for the indicator of environmental quality, we have selected ecological footprint. There is a long-lasting discussion about the selection of ecological footprint as an ecological indicator (Templet 2000; Kissinger and Haim 2008; Wiedmann and Barrett 2010; Moore et al. 2013; Galli 2015). Ecological footprint explains the carrying capacity of the collective area of the ecosystem to generate the resources which are consumed by the economic process, and also shows the ability to absorb the generated waste (Wackernagel et al. 1999; Solis-Guzman and Marrero 2015; Peng et al. 2019). There are two main reasons for selecting ecological footprint as an indicator of environmental quality. First, it represents the carrying capacity of the earth, and hence, a better measurement device for the sustainability of ecological system with compare to pollutants. According to Sustainable Development Goals (SDGs), the ecological footprint is the only tool which considers the ecological and biological capability of earth for sustainable economic activities and development (Rashid et al. 2018; Nazar et al. 2018; Pan et al. 2019). Second, natural resources (water, minerals, land resources, and forests) are consumed due to economic activities. So, if emissions are chosen as environmental indicators, then they will ignore the ecological sustainability as well as the significant human and industrial activities which can damage the sustainability of the ecosystem (Li et al. 2019).

Trade openness affects environmental quality negatively or positively by various channels according to the level of industrialization and globalization in an economy (Raza et al. 2017; Destek and Sinha 2020). According to the pollution haven hypothesis (PHH), a host economy with lax or slacker environmental laws becomes dirtier with trade openness (Copeland and Taylor 1994; Talukdar and Meisner 2001; Hoffman et al. 2005; He 2006; Baek and Koo 2009). Grossman and Krueger (1991) and Antweiler et al. (2001) made a holistic attempt at explaining the effects of trade openness on the environmental quality and decomposed these effects into scale effect, technique effect, and composition effect. The scale effect defines the expansion of the economy due to trade openness. With the consumption of natural resources and energy, the total amount of pollution generated will increase in the economy. The technique effect refers to the improvement of environmental quality as income level increases (Grossman and Krueger 1991). The composition effect of trade openness reflects a change in the constituents of a country’s output when it opens up to trade and becomes more globalized. If comparative advantages of a country are disposed to clean industries, then trade openness will lead to a shift from polluted or dirty goods to less polluted or clean goods and services. It is observed that the production process of dirty goods is capital-intensive while the output of clean goods is more human capital-intensive or labor-intensive. If the scale of production and other factors remain constant, a country that shifts its output towards capital-intensive products will create more pollution, while a country that shifts its production towards labor-intensive products or away from capital-intensive products will generate less pollution (Grossman and Krueger 1991; Cole and Elliott 2003; Shao et al. 2019).

Another concept comes into view about environmental quality is foreign direct investment (FDI), which enhances the growth process by the accumulation of capital, increasing productivity, and the diffusion of technology (Ali et al. 2012; Sarwar et al. 2013; Seker et al. 2015; Alvarado et al. 2017). The debate about the importance of FDI for environmental quality, which started in the early 1990s (UNCTAD 2008), has given two justifications about the relationship between FDI and environmental quality. First, FDI can cause the implementation of PHH, which explains that environmental regulations of an economy can impact the location of industries (Zarsky 1999; Millimet and Roy 2016; Sapkota and Bastola 2017). Chandran and Tang (2013), D’Agostino (2015), Sun et al. (2017), Solarin et al. (2017), and You and Lv (2018) confirm this hypothesis and describe that developed countries shift their industries to underdeveloped economies due to their lax environmental laws and regulations. Second, FDI can cause Pollution Halo Hypothesis, which advocates that foreign firms bring energy-efficient, advanced, and cleaner technology in the form of FDI to the host economy which improves environmental quality (Zarsky 1999; Wang et al. 2013). Therefore, FDI is considered as a determinant of degradation as well as improvement of environmental quality, although its relationship with the environment stays vague and needs further investigation (Al-Mulali and Tang 2013).

The relationship between institutional performance and the environmental quality is another complex topic. Institutional performance involves a complicated structure passing through different institutional channels and affects both political and market forces (Kang and He 2018; Glicksman et al. 2019). Designing targeted environmental and economic policies is assumed an essential role in encouraging the transition process but will need to be combined with improved institutions to guarantee that policies are monitored and implemented effectively (Dasgupta et al. 2016; Liao et al. 2017; Khan et al. 2019a, b). Therefore, the effectiveness of environmental policies depends on the policy adoption as well as on the institutional performance, cultural discourses, dominant ideas, the distribution of resources, and the industrial structure (Meyer et al. 2003; Hughes and Lipscy 2013). Politically and economically open societies which respect laws and regulations, market allocation of resources, and private property developed faster than those economies where these freedoms are restricted (Scully 1988; Knack and Keefer 1995; Hayek 2018).

The Organization of Islamic Cooperation (OIC) is an international organization grouping 57 countries, having Muslim population and constitutes a substantial part of the developing economies. OIC member states have collective population of over 1.8 billion people accounted for 23.8% of the total world population in 2017 (SESRIC 2018). In recent decades, due to industrial development, the environmental challenges in OIC countries have been accelerated (including water scarcity, climate change, land degradation, air pollution, and degradation of the coastal and marine ecosystem), fascinating considerable attention from international institutions (SESRIC 2017; Konac 2004). The environment has been defined as one of the ten areas of cooperation in the OIC plan of action of 2014 (COMCEC 2014). Due to trade openness and FDI, the volume of trade and production in OIC countries is expanding, so the consumption of natural resources and energy is also increasing which creates pollution (Sharif and Raza 2016; Mirjalili and Motaghian Fard 2019; Sharif et al. 2020). Among OIC countries, Indonesia recorded the highest total ecological footprint of 441 million global hectares followed by Turkey, Iran, and Nigeria with 267, 256, and 202 million global hectares of ecological footprint, respectively. On the other side, if we look on per capita ecological footprint, then Qatar has highest level of 14.4 global hectors of per capita ecological footprint followed by UAE, Bahrain, and Kuwait with 8.9, 8.6, and 8.6 global hectors of per capita ecological footprint as shown in Appendix A Table 10 (Global Footprint Network 2016). “Paris agreement” is considered the latest international deal and global initiative on climate change with 196 signatories, of which 166 have ratified it. Forty-three OIC member countries have ratified it, whereas 13 countries have the status of signatory (SESRIC 2018). Moreover, environmental scarcity and its degradation can generate economic and social conflicts in OIC countries (Tolba and Saab 2008; Nekooei et al. 2015; Gholipour and Farzanegan 2018).

Motivated by the ensuing contradiction in the literature, this study seeks to find out the overall effect of trade openness, FDI, and institutional performance on the environmental quality in OIC countries. Although many studies found this impact in different groups of economies, but in case of OIC countries, activities in this critical area are still very limited and comprehensive studies on the subject are even missing (Konac 2004; Pal and Tok 2019; Kaminski 2019). This study contributes to the literature in various ways: (i) this is the first study which examines the link between trade openness, FDI, institutional performance, and environmental quality in OIC countries; (ii) previous studies examined this relationship by using simple techniques, whereas, in this study, we have used new methodology named Dynamic Common Correlated Effects (DCCE), this technique overcomes the methodological issues which the existing literature suffers from; (iii) in this study, the use of ecological footprint as a new variable is an attempt to address the environmental issue in OIC countries in a modern context; (iv) this study gives valuable implications based on results, which will pursue the attention of policymakers and also make a significant contribution to the existing literature.

Literature review

There have been an increasing number of studies defining the impact of macroeconomic variables on environmental quality since last few decades. Many indicators of environmental quality are used and most are considered negative indicators, like pollutants. For example, Cole (2007) and Liao et al. (2017) used sulfur dioxide (SO₂), while Cole et al. (1997) used nitrous oxide (NO₂), and Abdouli and Hammami (2017), Zhang et al. (2017a, b), and You and Lv (2018) used CO₂ emissions. Very limited studies are found in which ecological footprint has been used, which is a positive indicator of environmental quality (Al-Mulali et al. 2015; Ozturk et al. 2016; Sharif et al. 2019). In this section, we will give a brief review of the literature and thereby, substantiating the parameterization of the study.

Various studies about environmental degradation considered trade openness as one of the important indicators that influence environmental quality (Grossman and Krueger 1991; Dinda 2004; Frankel 2009; Nekooei et al. 2015; Abdouli and Hammami 2017). Pioneer study of Grossman and Krueger (1991) investigated the impact of trade openness on the environment and found scale effect while taking composition effect and technique effect constant, and found that the impact of comparative advantage on environmental quality was conditional on environment and trade policies of a country (Copeland and Taylor 2004). Frankel and Rose (2005) analyzed the influence of trade openness on the environment with the help of instrumental variables and gravity model. The coefficients of openness and policy variable were found negative. In another study, Managi et al. (2009) investigated the impact of trade openness on environmental quality and found that trade openness had a positive effect on the environment in OECD countries. Similarly, Tsurumi and Managi (2014) explored the role of trade openness in the deforestation process in 142 countries and found that deforestation was slowed down in OECD countries and increased in non-OECD countries. Al-Mulali et al. (2015) found the effect of trade openness on the environment by using the ecological footprint for 58 selected countries. After applying difference and system GMM approaches, a positive impact was observed between trade openness and ecological footprint.

Nekooei et al. (2015) studied the association between trade openness, economic growth, democracy, and environment quality in OIC countries. Estimated results showed that trade openness, democracy, and growth had a positive effect on the environment. However, population had a negative and significant effect on the environment. Similarly, Ozturk et al. (2016) observed the impact of trade openness and urbanization on the ecological footprint for 144 countries for 20 years. Following the EKC framework, a positive relationship was found between trade openness and ecological footprint. Similarly, another study by Abdouli and Hammami (2017) explored the impact of FDI inflows, economic growth, and trade openness on environmental degradation for 17 MENA countries and confirmed the existence of EKC and pollution haven hypothesis in these countries. Other studies by Zhang et al. (2017a, b) observed negative relationship between trade openness and carbon dioxide emissions in 10 newly industrialized countries. You and Lv (2018) reported the spatial impact of economic globalization on pollution for the panel of 83 countries by applying spatial panel data method. There was negative indirect effects of economic globalization on CO₂ emissions and overcame the direct positive effects, and hence the total net effect was negative and significant. Similarly, Sharif et al. (2019) identified the impact of globalization on ecological footprint in top 10 globalized countries of the world by applying quantile-on-quantile approach. The results suggested that globalization had positive effect on ecological footprint in Denmark, Netherlands, Belgium, Switzerland, Norway, Portugal, Canada, and Sweden. On the other hand, there was a negative association between globalization and ecological footprint in the UK, Germany, France, and Hungary.

The impact of FDI on environmental quality is also discussed in number of empirical studies. Zhang (2011) elaborated the role of financial development in environmental degradation in China and indicated that FDI was mostly used in carbon-intensive production techniques and promoted CO₂ emissions. The validation of the pollution haven hypothesis was checked by Asghari (2013) in MENA countries by applying the fixed and random effect method. The outcomes indicated that FDI inflows had improved the environmental quality of MENA countries and also validated the pollution haven hypothesis. Wang et al. (2013) assessed both negative and positive aspects of FDI and analyzed that FDI increased the labor productivity, economic growth, and innovation in host city, but it also caused pollution and unemployment. In another study, Seker et al. (2015) found the effect of FDI and GDP on CO₂ emissions in Turkey over the period 1974–2010. The results indicated a positive and relatively small impact of FDI on CO₂ emissions, while the impact of GDP on CO₂ emissions was quite large. Similarly, Sun et al. (2017) validated the impact of trade openness and FDI inflows on CO₂ emissions by using ARDL model and observed that both trade openness and FDI promoted CO₂ emissions and the evidence of pollution haven hypothesis existed. Solarin and Al-Mulali (2018) looked at the influence of FDI inflows on carbon footprints, ecological footprints, and CO₂ emissions for 20 countries. Augmented mean group (AMG) estimation and common correlated effect (CCE) estimator were utilized. It was found that FDI inflows and urbanization mitigated pollution in developed countries while they increased pollution in the underdeveloped countries. Liu et al. (2018) analyzed the dynamics at work and spatial agglomeration effects in environmental pollution and FDI for the Chinese economy and concluded that FDI had a negative impact on different types of environmental pollutants in China. Similarly, Zafar et al. (2019) explored the effects of FDI on the ecological footprint using USA data from 1970 to 2015. ARDL model along Zivot-Andrews unit root test was used and the findings suggested that FDI had a negative association with the ecological footprint.

There are many studies that have linked various indicators of institutional performance with environmental quality by applying different econometric techniques. As pioneer studies, Torras and Boyce (1998) and Deacon (1999) found that democracy and good governance had a positive effect on environmental quality. This relationship is further supported by many studies including Neumayer (2003), Deacon (2003), Fredriksson and Svensson (2003), and De Mesquita et al. (2005). Bhattarai and Hammig (2001) examined the relationship between institutions and deforestation in 66 countries of Africa, Latin America, and Asia and found that good governance and political institutions had significantly reduced deforestation. Cole (2007) analyzed the linkage between corruption and pollution for 94 countries. Corruption had a positive direct impact on both CO₂ emissions and SO₂ emissions. Castiglione et al. (2012, 2013) observed that stronger rule of law had a negative association with pollution. Similarly, Liao et al. (2017) examined the nexus of income, energy, and SO₂ emissions for China by taking corruption as an institutional variable. The empirical evidence showed negative relationship between the number of anti-corruption cases and SO₂ emissions. Moreover, growth in income had a negative relationship with SO₂ emissions. Nekooei et al. (2015) explored the association between democracy and environment in selected OIC countries and showed a positive correlation between democracy and environment quality. Charfeddine and Mrabet (2017) analyzed the impacts of social and political factors and energy consumption on the ecological footprint for 15 MENA countries by applying DOLS and FMOLS approaches and observed that political institutions and energy consumption degraded environment by increasing the level of the ecological footprint. Gholipour and Farzanegan (2018) explored the role of governance in environmental protection in MENA countries. Results showed that GEEP (government consumption on environmental protection) alone could not promote environmental quality, but the rule of law and good governance were the main determinants of the final effects of GEEP on environmental quality.

Data and methodology

In this study, the dynamic common correlated effects of trade openness, FDI, and institutional performance on environmental quality are presented for OIC countries. From 57 OIC member countries, 47 countries are selected for sample size according to the availability of data (a list of selected countries is given in the Appendix B Table 11) for the period 1991–2016. Conventionally different techniques have been used in the estimation of panel data like GMM, random effect and fixed effect methods, but these techniques consider homogeneity and only allow changing the intercepts of cross-sectional units while, in the real world, the issue of heterogeneity in panel data exists. Therefore, nowadays the researchers around the world are more concerned about the cross-sectional dependence.

Various studies in the literature found that many times, panel data suffer from the problem of cross-sectional dependence. There are many unobserved factors and shocks which occur at the same time due to financial or economic integration of countries (De Hoyos and Sarafidis 2006; Dogan et al. 2017; Latif et al. 2018). Therefore it is necessary to check whether all cross-sectional units are equally affected by shocks. Contemporary methods of estimation have gained remarkable worldwide attention in the field of macroeconomic research, but during this era of globalization where countries suffer a lot from changes in some other countries, it is impossible to proceed with traditional methods. For this purpose, the most recent DCCE approach by Chudik and Pesaran (2015) is applied.

The DCCE approach is designed on the principles of pooled mean group (PMG) technique developed by Pesaran et al. (1996), Mean group (MG) estimation developed by Pesaran and Smith (1995), and CCE estimation developed by Pesaran (2006). This approach considers main issues which are not recognized by other conventional methods. The most important is the consideration of the cross-sectional dependence by taking the logs and averages of cross-sectional units. Secondly, it allows for heterogeneous slopes and dynamic common correlated effects. One of the other implications of this technique is that it is also applicable if the sample size is small by using the method of Jackknife correction (Chudik and Pesaran 2015). This approach can also be used in case of unbalanced panel data (Ditzen 2016) and when structural breaks are present in data (Kapetanios et al. 2011). The equation of DCCE model can be written as follows:

$$ LNEC{F}_{it}={\alpha}_i LNEC{F}_{it-1}+{\delta}_i{X}_{it}+\sum \limits_{p=0}^{p_T}{\gamma}_{xip}{\overline{X}}_{t-p}+\sum \limits_{p=0}^{p_T}{\gamma}_{yip}{\overline{X}}_{t-p}+{\mu}_{it} $$

In this equation, LNECF shows a log of ecological footprint and its lag is used as an independent variable and X_it shows a set of other independent variables, and P_T represents lag of cross-sectional averages.

One of the major problems with previous studies is that institutional performance is proxied by a single variable measures such as government stability (Naude and Saayman 2005; Ingram et al. 2013; Habibi 2017), corruption (Saha and Yap 2015; Lv and Xu 2017; Meo et al. 2018), and law and order (Moyo and Ziramba 2013; Cui et al. 2016; Gozgor et al. 2019). Use of a single proxy for institutional performance could be biased and misleading. Moreover, indicators of institutional performance are highly associated with each other (Globerman and Shapiro 2002; Daude and Stein 2007; Buchanan et al. 2012; Sabir et al. 2019), and it is difficult to combine all the indicators in one equation (Ullah and Khan 2017; Sabir et al. 2019). Therefore, we have used panel principal component analysis (PCA)¹ estimation technique to obtain the variable of institutional performance (IP) made up of five selected indicators (socioeconomic conditions, government stability, law and order, corruption, and investment profile). These indicators are chosen on the basis of their ability to affect environmental quality as explained by various studies (Habibi 2017; Liao et al. 2017; Meo et al. 2018; Gholipour and Farzanegan 2018; Gozgor et al. 2019). These indicators reflect various institutional factors and important issues that significantly affect the environment and ecosystem of a country (Bhattarai and Hammig 2001; Mavragani et al. 2016; Charfeddine and Mrabet 2017). The IP duplicates the original data of all the indicators of institutions into a single variable with minimal loss of information (Tang and Tan 2014; Le et al. 2016; Batuo et al. 2018; Khan et al. 2019a, b). The ICRG assigns a rating to these indicators on the scale of 100 points, the highest score indicates low risk (high quality), and the lowest score is the evidence of high risk (low quality).² We prefer the indicators of International Country Risk Guide (ICRG) to the Worldwide Governance Indicators (WGI) due to their longitudinal coverage.³

In panel PCA technique, the jth factor index can be written as:

$$ I{P}_j={W}_{j1}{X}_1+{W}_{j2}{X}_2+{W}_{j3}{X}_3+\dots \dots +{W}_{jp}{X}_p $$

Here, IP_j is the institutional performance; W_j represents weight of the parameter of the factor score; original figure of the respective indicators are represented by X; while P shows the number of variables in the equation.

Many previous studies like Maddala and Wu (1999), Levin et al. (2002), and Im et al. (2003) used first-generation unit root tests which have not considered cross-sectional dependence. Therefore, a second-generation unit root test is also used in this study which is developed by Pesaran (2007) and Choi (2006) and provides more reliable results. The null hypothesis of no cross-sectional dependence is tested against the alternative hypothesis of cross-sectional dependence. To estimate long-run results, a bootstrap cointegration technique by Westerlund and Edgerton (2008) is employed as compared to conventional cointegration techniques. This approach is preferred because it considers structural breaks and in short time dimension, the outcomes are more sensitive to lag and lead length (Persyn and Westerlund 2008). The variables of this study are presented in Table 1 with a detail description of the unit of measurement and sources of data collection.

Table 1 Description of variables

Results and discussion

Descriptive statistics (measures of dispersion and central tendency) and pair-wise correlation for all the variables (in log form) are presented in Table 2. Ecological footprint is significantly correlated with all the explanatory variables (i.e., trade openness, institutional performance, FDI, and urbanization).

Table 2 Descriptive statistics and pair-wise correlation

Due to the nature of macroeconomic variables and panel data, there are increasing chances that cross-sectional dependence may prevail. For this purpose, Pesaran (2004)-CD, Pesaran (2004) scaled LM, and Baltagi et al. (2012) biased-corrected scaled LM tests are employed to check the presence of cross-sectional dependence as shown in Table 3. Findings of these tests are useful not only to decide estimation technique but also helpful to decide that whether first-generation panel unit root tests (Levin et al. 2002; Im et al. 2003) are suitable which assume cross-sectional independence or second-generation panel unit root tests (Chang 2004; Pesaran 2007) are more appropriate that consider the cross-sectional dependence. In this study, both types of panel unit root tests are conducted to avoid misleading results.

Table 3 Results of cross-sectional dependence tests

The null hypothesis for above cross-sectional dependence tests is the absence of cross-sectional dependence, and according to the empirical findings of this study, there is sufficient evidence to reject the null hypothesis and conclude that there is a presence of cross-sectional dependence among the cross-sectional units.

Results of first-generation panel unit root tests are presented in Table 4. Variables are stationary at the level and first difference; however, no variable is stationary at second difference.

Table 4 First-generation panel unit root tests

In Table 5, the findings of the second-generation panel unit root test (Pesaran 2007) are presented, which is also called CIPS-Test. This test considers the cross-sectional dependence among the variables, and according to these outcomes, the variables have again mix order of stationarity at the level and first difference and no one is stationary at the second difference.

Table 5 Results of CIPS-Test

Pedroni (1999) test of cointegration is applied to check the long-run association among the variables, as shown in Table 6. There are seven statistics which are divided into sections within and between dimensions, and according to these statistics there is no close evidence of long-run association among the variables. The results of Pedroni (1999) test may be misleading because it does not consider several important issues like heteroskedasticity, serial correlation, structural breaks, and cross-sectional dependence among the countries or cross-sectional units, while Westerlund and Edgerton (2008) is an advanced test of cointegration among the variables because it considers all above-mentioned issues, therefore, the results are more reliable. Outcomes of Westerlund test (Table 7) shows that probability values of Ga, Gt, Pa, and Pt of Persyn and Westerlund (2008)⁴ cointegration tests are significant, which reject the null hypothesis of no cointegration and confirm the presence of long-run relationship among the variables.

Table 6 Pedroni residual cointegration tests

Table 7 Westerlund ECM panel cointegration test

Estimation of the PMG is presented in Table 8, which rejects the null hypothesis of no cointegration and shows a positive relationship of trade openness, FDI, and institutional performance with the ecological footprint. According to these estimates, all the relationships are according to expectations except the relationship between institutional performance and ecological footprint, which is surprising. It can be the result of flaws in the earlier approaches like MG, PMG, and AMG estimators. In other words, a change affecting one country will be transferred to other countries due to trade openness and globalization, and therefore, DCCE is the most appropriate estimation technique in this situation. To deal with the flaws of earlier estimation techniques, the primary purpose of this study is to apply DCCE estimation technique, and the results are presented in Table 9 which rejects the null hypothesis of no cointegration.

Table 8 Results of PMG estimation

Table 9 Results of DCCE estimation

Results show a positive association of trade openness, FDI, and urbanization with ecological footprint, which means any increase in these variables will degrade the environmental quality. A positive association between trade openness and ecological footprint in DCCE estimation also confirms the existence of PHH in OIC countries according to which host economies with slacker environmental regulations becomes dirtier with trade openness (Copeland and Taylor 1994). The finding is also consistent with the studies of Talukdar and Meisner (2001), Xing and Kolstad (2002), Dinda (2004), Hoffman et al. (2005), He (2006), and Baek and Koo (2009).

The positive association between FDI and ecological footprint in DCCE estimation is also explained by PHH which describes that due to FDI, developed countries shift their production to underdeveloped countries where environmental regulations are weak. The finding is also consistent with the studies of Chandran and Tang (2013), D’Agostino (2015), Sun et al. (2017), Solarin et al. (2017), and You and Lv (2018). The institutional performance has a negative and significant relationship with the ecological footprint. It means improvements in determinants of institutional performance, i.e., socioeconomic conditions of a country, the stability of government, corruption, and law and order, will reduce the ecological footprint and hence improve the environmental quality of OIC countries. The finding is aligning with Bhattarai and Hammig (2001), Cole (2007), Nekooei et al. (2015), and Liao et al. (2017). Furthermore, we have found a significant and positive relationship between urbanization and ecological footprint, which means an increase in urban population will cause environmental degradation in OIC countries.

Thus we have realized that PMG estimation provides some misleading results (as the unexpected relationship of institutional performance and ecological footprint shown in Table 8). Moreover, Neal (2015) and Arain et al. (2019) found that PMG, MG, DOLS, and FMOLS methods provide conflicting results in the presence of cross-sectional dependence. So, it is better to rely on the results of DCCE estimation technique in the presence of cross-sectional dependence.

Concluding remarks and recommendations

The present study has evaluated the dynamic common correlated effects of trade openness, FDI, and institutional performance on environmental quality in OIC countries by taking the data for the year 1990 to 2016. We have used newly developed DCCE approach to tackle the weaknesses of earlier methods of panel data such as PMG, AMG, and MG estimators. It is also found that trade openness and institutional performance have more substantial effects on ecological footprint than FDI. In our study, it is found that FDI and trade openness are the cause of environmental degradation in OIC countries. Additionally, this relationship also relies on the negative externalities that can be created through natural demand in the form of environmental degradation. The results concluded that trade openness and FDI inflows in these countries harm the nature and ecological system instead of focusing on proper utilization of technology-oriented resources. OIC countries should shift their production from labor-intensive to capital-intensive technology because capital-intensive technology is more appropriate for clean production technology. Policymakers should prefer green investment, green technology, and low-carbon production. To achieve sustainable development goals (SDGs) like better health, low-cost and green energy, infrastructure, openness, responsible consumption and production, and climate change, the green technology is the appropriate option; it will reduce ecological footprint when compared to conventional production technology.

Index for institutional performance is developed through panel PCA by taking five indicators of institutional quality or performance. However, for future research, the indicators of institutional performance like socioeconomic conditions, government stability, corruption, and law and order can be treated separately for the individual effect of each indicator on ecological footprint. Furthermore, we took overall ecological footprint in this study, while for the future research, ecological footprint can be segregated in different parts, i.e., carbon footprint, biocapacity, cropland, grazing lands, fishing grounds and forest products, etc. The rapid increase in urban population will result in the growth of squatter settlements and urban slums and the overburdening of the water supply and waste-disposal systems, thus resulting in environmental degradation. Government of OIC countries should make different productive projects and produce new jobs in villages so that migration towards big cities should be minimized. The main reservation from the analysis pertains that whether society prevails to live in the way to affect the environmental quality or whether they can change the production methods which can upgrade towards clean production which leads to a sustainable environment by outweighing the adverse effects of trade openness and FDI. The insistence on sustainable environment is required to check out how the economies are utilizing their resources and natural capacity. Knowing about the aspects of what factors can cause positive effects on natural accounts and what causes the declination of natural reserves will lead to better awareness of the potentiality of business, production, and environmental quality. Moreover, the proper management and utilization of natural resources will lead to achieve the target of green economy and improved environmental quality.

Appendices

Appendix A

Table 10 List of OIC countries with ecological footprint

Appendix B

Table 11 List of selected OIC countries in sample