The Fiscal and Monetary Policies and Environment in GCC Countries: Analysis of Territory and Consumption-Based CO₂ Emissions

Abstract

Expansionary monetary and fiscal policies are necessary for economic and environmental development. The present research studies the impact of monetary policy and fiscal policy on Territory-Based CO₂ (TBC) and Consumption-Based CO₂ (CBC) emissions in Gulf Cooperation Council (GCC) economies from 1990–2019. The cointegration is corroborated through various tests, and long-term relationships are found in both TBC and CBC models. Government expenditures have long-term positive effects on both TBC and CBC emissions and short-term positive effects on TBC emissions in the region. Money supply negatively affects the TBC and CBC emissions in the long run and positively affects TBC and CBC emissions in the short run. Hence, monetary policy needs a long time to have positive ecological effects in the GCC region. Moreover, fiscal policy in both the long and short run and monetary policy in the short run have scale effects in GCC economies. Therefore, we recommend reducing fiscal measures and encouraging monetary policy in the long run to have positive environmental outcomes in the region.

1. Introduction

Grossman and Krueger [1] have initiated the idea of linking economic growth and pollution emissions. The economy may be tracked into pollution emissions activities at the first phase of growth [2], which is a scale effect. Onwards, an economy may improve environmental health due to demand for a cleaner environment and technical and composition changes using renewable sources [3,4]. In both stages, the macroeconomic policies would play their role in greening the economy. On the other hand, macroeconomic policies would also adversely affect the environment if the primary object were just economic growth. For example, expansionary monetary policies in terms of increasing money supply and decreasing bank rate would increase the investment, industrialization, and aggregate demand in the economy [5], which would increase energy demand and pollute the economy if the energy mix is relying primarily on the fossil fuels [6]. Alternatively, literature suggests using a green monetary policy to reduce the environmental effects of the expansionary policy. Wang and Zhi [7] argued that green finance with adequate policies and market mechanisms would be helpful for ecological balance. He et al. [8] explored this issue in China and found that green finance improved the investment in renewable projects to some extent but has a negative effect on the overall loan market. Hence, there is a need for intensive government regulations to achieve the optimum results from green financing in the economy. On the other hand, Schoenmaker [9] claimed that any monetary policy effort to promote a green economy might be nullified with the environmental consequences of usual expansionary policies. In the presence of both positive and negative ecological consequences in the literature, it seems pertinent to test the exact linkages between monetary policy and pollution emissions to conclude the precise effect of monetary policy on the environment in any economy or region.

In addition to monetary policy, the literature has suggested that globalization and oil prices could asymmetrically affect pollution emissions [10,11]. Moreover, Weimin et al. [12] found that innovation and globalization helped to improve the environmental quality. Further, Leal and Marques [13] provided evidence that higher globalization helped to shape the Environmental Kuznets Curve (EKC). On the other hand, Pata and Caglar [14] reported that globalization accelerated pollution, and the EKC could not be found in China. In another dimension, Rehman et al. [15] investigated and found that the agriculture sector and rainfall positively affected CO₂ emissions. Janjua et al. [16] reported that foreign direct investment (FDI) and the industry sector were responsible for pollution. On the other hand, the literature found a mix of the negative and statistically insignificant effects of FDI on CO₂ emissions [17,18]. Khan et al. [19] found that fossil fuels reduced and renewable energy improved the environmental quality in Europe. Moreover, bulk literature has suggested that the concept of a circular economy and green business practices would serve to promote eco-environmental performance [20,21,22,23,24,25,26].

In terms of fiscal policy, fiscal instruments also significantly affect the aggregate demand in the economy, directly and indirectly [27], which impacts the environment through industrialization, economy size, and energy usage. Moreover, environmental policies could discourage investment and other aggregate demand components [28]. Furthermore, environmental taxes can reduce the environmental side effects [29]. Ulucak et al. [30] narrated that environmental tax with high globalization could help in mitigating CO₂ emissions. In the same way, technology and patents supported by government policies would also help in reducing CO₂ emissions. Gaspar et al. [31] argued that pollution is a threat to the whole world, which may be reduced by carbon taxation on coal and other fossil fuel consumption. It would help the economies to shift towards cleaner energies. Further, the authors explained that USD 75 should be imposed on every ton of CO₂ emissions to achieve the target to reduce global warming by 2 °C in 2030. Contrarily, government spending could increase the dimension and causes of pollution emissions [32]. For example, government spending may increase economic activities and emissions through the scale effect. On the other hand, government spending in health and education may increase the income, human capital, and awareness that individuals have of a cleaner environment. It can also enhance the administrative and institutional ability of the government for environmental protection [33]. In this regard, empirical evidence has been provided showing that government expenditure has been found helpful in controlling consumption and production-based emissions [34]. Moreover, Ullah et al. [35] investigated the asymmetrical effects of fiscal variables on environmental quality and found mixed evidence. For instance, increasing government expenditures reduced the environmental quality in some investigated countries and improved it in others. The inverse effects were reported for decreasing government expenditures.

Following Halkos and Paizanos’s [34] arguments about the role of fiscal policy on Territory-Based CO₂ (TBC) and Consumption-Based CO₂ (CBC) emissions, we are highly motivated to estimate the effect of macroeconomic policies on such emissions. The literature has suggested the analysis of TBC and CBC emissions to differentiate the emissions in consumption and production activities [36]. CBC emissions are adjusted from TBC emissions by adding import emissions and subtracting export emissions. Hence, exports would reduce and imports would increase CBC emissions [37]. Following this idea, Mahmood [38] did an analysis and found exports reduced and imports increased CBC emissions. Hence, the impact of exports was negative, and the effect of imports was positive on CBC emissions. Khan et al. [39] and Hasanov et al. [36] also did the same analysis and found the same conclusions. Moreover, the industrial sector contributed to CBC emissions. In this context, no study focuses on the role of macroeconomic policy on TBC and CBC emissions. Furthermore, the testing of combined fiscal and monetary policies on CO₂ emissions is rare in global literature [40,41,42,43] and missing in the Gulf Cooperation Council (GCC) region. The analysis of the GCC region is critical as emissions are rising at a higher projected pace [44]. It may be due to the expansionary fiscal policy of GCC countries. This is because GCC economies’ government revenues largely depend on the oil sector and increasing the oil sector has both direct and indirect effects on pollution emissions. By direct impact, the oil sector emits a lot of pollution due to oil production activities. By indirect impact, the oil sector supports the economic growth in GCC economies. Resultantly, the booming oil sector enables the GCC governments to spend more in the economies, increasing the pollution emissions due to the scale effect of expansionary fiscal policy and increasing overall economic activities due to increasing income and aggregate demand. Moreover, expansionary monetary policy would also have a scale effect on energy consumption and pollution emissions. Therefore, it is pertinent to analyze which fiscal or monetary policy has more significant environmental consequences to suggest the right policy choice for demanding a better environment for sustainable development goals in GCC countries. Hence, the present research estimates the effect of money supply and government expenditures on the CBC and TBC emissions to ensure an empirical contribution in global and GCC literature and proposes the right policy for economic and environmental sustainability in GCC economies.

The research is distributed into four major sections. The first section provides a detailed literature review. The second section delivers a theory to draw a contextual picture of the scenarios and model being discussed and also covers the data and methodology. The third section provides an analysis. Lastly, the fourth section delivers concluding remarks with policy implications and future directions.

2. Literature Review

Fiscal policy could directly affect the environment, and literature has investigated this issue in past studies. For instance, Neves et al. [45] examined 17 European Union (EU) economies from 1995–2017 and found that environmental regulations, renewable energy policies, and Foreign Direct Investment (FDI) reduced emissions. Emissions taxes and FDI have promoted the environmental profile of the EU. Moreover, the pollution halo hypothesis has been corroborated in the EU. Le and Ozturk [46] argued that expansionary fiscal policy would be helpful to increase the capacity to buy renewable technology and green products. The authors worked on 47 emerging economies from 1990–2014. The authors found that the financial sector and globalization accelerated CO₂ emissions and recommended the role of financing and governing economic sectors for sustainable development. Ahmed et al. [47] explained that government spending towards clean energy infrastructure might reduce CO₂ emissions. The authors worked investigated Japan from 1974–2017 and found that government spending towards clean energy infrastructure and nuclear energy reduced CO₂ emissions. Zhang et al. [48] argued that public support for green projects might promote green technologies and reduce emissions. They found that public spending on clean technology research promoted sustainable development. Moreover, this type of policy would accelerate renewable consumption through green technology innovation, which facilitates a reduction in pollution emissions.

Yilanci and Pata [49] investigated G7 economies from 1875–2016 by bootstrap causality test and found that public spending helped reduce emissions. However, they could not validate the EKC with a long-term analysis. The same finding has been shared by Katircioglu and Katircioglu [50] in the case of Turkey from 1960–2013 where public spending as a share of income reduced the emissions. Moreover, the EKC was confirmed in Turkey. Hashmi and Alam [51] found that environmental taxes reduced CO₂ emissions. Moreover, green patents also reduce CO₂ emissions, which have a more significant effect than environmental taxes. Aydin and Esen [52] investigated a nonlinear effect of energy, transport, and pollution taxes in the EU from 1995–2013. They reported a significant positive impact of transport tax on emissions in the first phase of EKC and then a negative effect in the second phase. However, an insignificant impact in the first phase and a negative impact in the second phase of EKC was found in the rest of taxes. Mardones and Flores [53] reported that carbon taxes helped accelerate a cleaner industrial use of energy in Chile and reduced emissions. However, low tax rates showed insignificant effects. Yuelan et al. [54] explored the impact of public revenues and spendings in China from 1980–2016. In the long run, they found that expansionary policy was responsible for increasing CO₂ emissions. Nevertheless, its short-term effects were environmentally pleasant. Solaymani [55] investigated the role of taxes on emissions in Malaysia and found that carbon and energy taxes helped reduce emissions. Moreover, a carbon tax is more efficient than an energy tax in reducing fossil fuel consumption and CO₂ emissions. Khan et al. [56] investigated and reported that carbon taxes, renewable energy, and innovations helped reduce carbon emissions.

Literature has investigated the environmental effects of monetary policy. Qingquan et al. [5] examined Asian economies from 1990–2014 and found that increasing money supply raised CO₂ emissions and decreasing money supply reduced emissions. Hence, expansionary measures increased emissions and vice versa for contractionary monetary policy. However, contractionary monetary policy would reduce the investments in technical innovation and may reduce the chance to produce cleaner technologies, which may have adverse environmental effects in the long run. Moreover, empirical findings show that human capital helped reduce emissions, and remittance and fossil fuel accelerated CO₂ emissions. Qingquan et al. [57] investigated quarterly data of Australia from 1972–2014. Expansionary commercial measures accelerated CO₂ emissions, and contractionary commercial and monetary efforts helped reduce CO₂ emissions. Moreover, remittances and fossil energy usage increased CO₂ emissions in Australia. Vo and Zaman [58] investigated 101 economies from 1995–2018 and found the EKC between income and emissions variables. Most economies were found on the first stage of EKC and had environmental consequences of growth. FDI and energy use accelerated CO₂ emissions. Moreover, the rising money supply increased the CO₂ emissions in a large sample of the world’s economies with a causal feedback effect. Furthermore, the money supply also caused FDI and increased income.

Limited studies probed the simultaneous effects of fiscal and monetary policies. For instance, Ullah et al. [40] argued that expansionary policy could accelerate energy demand and pollution in the short run. In their empirical exercise, the authors found that fiscal and monetary shocks showed adverse environmental effects in the short run and showed pleasant environmental impact in Pakistan in the long run. Therefore, monetary and fiscal policies need a long time to have a positive environmental effect. Chan [41] narrated that contractionary monetary policy, for example, by increasing bank rates, would reduce economic activities and aggregate demand. Hence, it would help mitigate CO₂ emissions but would reduce the general welfare of society as well. The author suggested that fiscal policy through carbon taxes can only achieve environmental targets and higher household welfare levels. Hence, carbon taxes should be complemented by a monetary policy.

Chishti et al. [59] investigated and found that expansionary (contractionary) fiscal measures increased (decreased) CO₂ emissions, while expansionary and contractionary monetary measures increased CO₂ emissions. Moreover, fossil use and consumption expenditure increased CO₂ emissions, and renewable usage reduced CO₂ emissions. Using the period from 1990–2019, Mughal et al. [42] found that increasing (decreasing) bank rates reduced (raised) CO₂ emissions. Moreover, rising public spending reduced CO₂ emissions. The same finding of both expansionary policies was reported in the short run, but contractionary policies carry an insignificant impact. Bhowmik et al. [43] investigated the US and found that monetary policy uncertainty accelerated CO₂ emissions and fiscal policy uncertainty reduced CO₂ emissions. However, trade policy uncertainty did not affect CO₂ emissions. Ahmed et al. [60] investigated G7 from 1985–2017 and found that environmental regulations and democracy helped reduce ecological footprints. However, income accelerated ecological footprints. Mahmood et al. [61] probed the effect of Financial Development (FD) on emissions and reported an insignificant effect of FD and a positive effect of income.

In the GCC context, Mahmood and Furqan [62] examined the impact of oil rents and other determinants on greenhouse gas emissions. Oil rents showed a significant effect on most of the investigated emissions. Moreover, FD increased and FDI decreased most of the emissions. Moreover, Mahmood et al. [63] corroborated the positive and asymmetrical impact of urbanization and industrialization on CO₂ emission in Saudi Arabia. Furthermore, oil price, stock market, and trade nexus have been investigated in the GCC context [64,65]. However, the existing GCC literature did not work on the macroeconomic policy and emissions relationship. Hence, this present research attempts to test the effects of monetary and fiscal policies on CBC and TBC emissions to fill this gap.

3. Methods

The economic policies play an influential role in determining the pollution emission in any country because any expansionary policy can increase the economic activities through a scale effect. The fiscal policy can have an immediate impact because government spending is a direct component of aggregate demand in an economy. Increasing government spending increases production and consumption activities and aggregate demand. Therefore, it would increase investment and production, which would increase the energy demand and pollution emissions if renewable sources are ignored in the energy mix of an economy. Moreover, expansionary monetary policy may also increase aggregate demand through rising consumption and investment activities. Hence, the expansionary monetary policy could also pollute the environment. Overall, any expansionary fiscal or monetary policy could have environmental consequences if the energy sector does not transition into cleaner sources in the process of expansionary policy. Literature has probed the combined environmental impact of both fiscal and monetary measures. Moreover, Halkos and Paizanos [27] suggested that macroeconomic policies would influence production and consumption-based emissions. Hence, we hypothesize the impact of both fiscal and monetary policies on TBC and CBC emissions in the following way:

LCBC_it = f (LGE_it, LMS_it)

(1)

LTBC_it = f (LGE_it, LMS_it)

(2)

LCBC_it is consumption-based and LTBC_it is tertiary-based CO₂ emissions. LGE_it is government expenditure, which is a proxy for fiscal policy. LMS_it is money supply, which is a proxy of monetary policy. i shows 6 GCC economies, and t shows a period of 1990–2019. All series are used in natural log form. CO₂ emissions are taken from Global Carbon Atlas [66], and policy variables are taken from the World Bank [67]. LGE_it could positively affect CO₂ emissions through the scale effect and might have a negative impact if government spending is encouraging cleaner technologies in an economy. In the same way, LMS_it could positively impact CO₂ emissions through the scale effect and might negatively impact if the monetary policy encourages the consumption of cleaner sources. For example, financial sector loans are facilitated in cleaner technology projects. The exact relationship would be found after a thorough empirical investigation. The relationship would be confirmed by applying panel cointegration. Literature suggests checking the integration level to verify the suitability of cointegration analyses [68,69]. For this purpose, three tests are used, which are Im-Pesaran-Shin (IPS), Levin-Lin-Shin (LLS), and Fisher-Augmented Dickey Fuller (ADF). These verify the robustness of each other, which are provided by Im et al. [70], Levin et al. [71], and Maddala and Wu [72]. Afterward, cointegration is used. Kao [73] provided the residual-based test and cointegration can be claimed if the Ordinary Least Square (OLS) residual proves to be at a stationary level. To test robustness, we applied Maddala and Wu’s [72] procedure to find the aggregate probability of the Johansen [74] statistics, which may confirm the cointegration through investigating the vectors using the following aggregating formulae:

$$y=-2{{\displaystyle \sum }}_{i=1}^{N}ln\left(probabilit{y}_i\right)$$

(3)

After testing cointegration through the above equation, the robustness may again be verified through the Pedroni [75] test with the following statistics:

$$T^2{N}^{1.5}{Z}_{\widehat{v} N,T}=T^2{N}^{1.5}{\left({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^{T}1/{\widehat{L}}_{11i}^2{\widehat{e}}_{i,t-1}^2\right)}^{-1}$$

(4)

$$T{N}^{0.5}{Z}_{\widehat{\rho }N,T-1}=T^2{N}^{0.5}\left({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^T\frac{1}{{\widehat{L}}_{11i}^2{\widehat{e}}_{i,t-1}\Delta }{\widehat{e}}_{i,t}-{\widehat{\lambda }}_i\right){\left({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^{T}1/{\widehat{L}}_{11i}^2{\widehat{e}}_{i,t-1}^2\right)}^{-1}$$

(5)

$$Z_{t N,T}=\left({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^{T}1/{\widehat{L}}_{11i}^2{\widehat{e}}_{i,t-1}\Delta {\widehat{e}}_{i,t}-{\widehat{\lambda }}_i\right) {(\tilde{\sigma }}_{N,T}^2{{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^T{\widehat{L}}_{11i}^{-2}{\widehat{e}}_{i,t-1}^2{)}^{-0.5}$$

(6)

$$Z_{t N,T}^{*}=\left({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^T\frac{1}{{\widehat{L}}_{11i}^2{\widehat{e}}_{i, t-1}^{*}}\Delta {\widehat{e}}_{i,t}^{*}\right){({\tilde{s}}_{N,T}^{*2}{{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^{T}1/{\widehat{L}}_{11i}^2{\widehat{e}}_{i,t-1}^{*2})}^{-0.5}$$

(7)

$$T{N}^{0.5}{\tilde{Z}}_{\widehat{\rho }N,T-1}=T.N^{-0.5}\left({{\displaystyle \sum }}_{t=1}^T{\widehat{e}}_{i,t-1}\mathsf{\Delta }{\widehat{e}}_{i,t}-{\widehat{\lambda }}_i\right){({{\displaystyle \sum }}_{i=1}^N{{\displaystyle \sum }}_{t=1}^T{\widehat{e}}_{i,t-1}^2)}^{-1}$$

(8)

$$N^{-0.5}{\tilde{Z}}_{t N,T}=N^{0.5}{\left({{\displaystyle \sum }}_{t=1}^T{\widehat{e}}_{i,t-1}\mathsf{\Delta }{\widehat{e}}_{i,t}-{\widehat{\lambda }}_i\right)}^{-1}{{\displaystyle \sum }}_{i=1}^N{\left({\widehat{\sigma }}_i^2{{\displaystyle \sum }}_{t=1}^T{\widehat{e}}_{i,t-1}^2\right)}^{-1}$$

(9)

$$N^{-0.5}{\tilde{Z}}_{t N,T}^{*}=N^{0.5}{\left({{\displaystyle \sum }}_{t=1}^T{\widehat{e}}_{i, t-1}^{*}\mathsf{\Delta }{\widehat{e}}_{i, t}^{*}\right)}^{-1}{{\displaystyle \sum }}_{i=1}^N{\left({{\displaystyle \sum }}_{t=1}^T{\widehat{s}}_i^{*2}{\widehat{e}}_{i,t-1}^{*2}\right)}^{-0.5}$$

(10)

Considering heterogeneous country effects, the above equations would suggest cointegration. Afterward, we applied the Pesaran et al. [76] method to find the effects of our policy variables on CO₂ emissions. This method was chosen as it is a Pooled Mean Group (PMG) and provides robust results in a mixed order of integration.

$$\mathsf{\Delta }{z}_{it}={\alpha }_i+{{\displaystyle \sum }}_{j=1}^{p-1} {\gamma }_j\mathsf{\Delta }{z}_{i, t-1}+{{\displaystyle \sum }}_{j=0}^{q-1}{\beta }_j\mathsf{\Delta }{w}_{i,t-1}+{\mu }_1{z}_{i, t-1}+{\mu }_2{w}_{i, t-1}+e_{1it}$$

(11)

$$\mathsf{\Delta }{z}_{it}={\alpha }_i+{{\displaystyle \sum }}_{j=1}^{p-1} {\gamma }_j\mathsf{\Delta }{z}_{i, t-1}+{{\displaystyle \sum }}_{j=0}^{q-1}{\beta }_j{\mathsf{\Delta }\mathrm{w}}_{i,t-1}+{\phi }_j{u}_{i,t-1}+e_{2it}$$

(12)

Equations (11) and (12) test the long- and short-term effects. Afterward, the robustness of PMG was tested by the Fully Modified OLS (FMOLS) of Pedroni [77] and Dynamic OLS (DOLS) of Kao and Chiang [78] in the following way:

$${\widehat{\beta }}_{FMOLS}=\left({{\displaystyle \sum }}_{n=1}^N\left({{\displaystyle \sum }}_{t=1}^T\left(w_{it}-{\overline{w}}_i\right){\widehat{z}}_{it}^{+}+T{\widehat{\Delta }}_{\epsilon \mu }^{+}\right)\right)/\left({{\displaystyle \sum }}_{t=1}^N{{\displaystyle \sum }}_{t=1}^T{\left(w_{it}-{\overline{w}}_i\right)}^{\prime }\right)$$

(13)

$${\widehat{\beta }}_{DOLS}=\left({{\displaystyle \sum }}_{t=1}^T{z}_{it}{\widehat{z}}_{it}^{+}\right)·{{\displaystyle \sum }}_{i=t}^{N}1/{{\displaystyle \sum }}_{t=1}^T{z}_{it}{z}_{it}^{\prime }$$

(14)

Equations (13) and (14) are the modified and dynamic versions of OLS and provide robust results by removing many econometric problems.

4. Data Analyses and Discussions

In

Table 1. Unit root analyses.

Series	IPS		LLC		Fisher-ADF
	Constant	Constant and Trend	Constant	Constant and Trend	Constant	Constant and Trend
Level
LCBCit	0.5759	−1.0772	−0.5565	−3.0406	12.0614	15.9109
	(0.7176)	(0.1407)	(0.2889)	(0.0012)	(0.4408)	(0.1953)
LTBCit	−1.2107	1.4581	−0.9775	0.4942	17.2302	4.8899
	(0.1130)	(0.9276)	(0.1742)	(0.6894)	(0.1411)	(0.9616)
LGEit	−0.9902	−1.6428	−1.7228	−1.8842	23.3242	23.9536
	(0.1611)	(0.0502)	(0.0425)	(0.0298)	(0.0245)	(0.0206)
LMSit	−3.6069	−4.9801	−4.5227	−8.2021	42.6335	82.3455
	(0.0002)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)
First Difference
ΔLCBCit	−6.1309	−5.0956	−4.2631	−3.0304	59.4586	47.0703
	(0.0000)	(0.0000)	(0.0000)	(0.0012)	(0.0000)	(0.0000)
ΔLTBCit	−8.2643	−6.9252	−6.3976	−4.6417	82.3608	63.2934
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)
ΔLGEit	−7.4991	−6.2432	−6.8013	−4.8588	74.7894	57.9564
	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)
ΔLMSit	−9.1672	−8.1078	−6.7832	−5.1972	93.0122	75.8042
	(0.000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)	(0.0000)

Note: (p-value).

, panel unit root results are shown. Leveled LCBC_it and LTBC_it have unit roots. However, LGE_it and LMS_it are stationary in most tests. Moreover, all differenced variables are stationary. Hence, the results corroborate a mixed integration order.

Table 2. Cointegration in LTBC_it model.

	Statistic	p-Value	Weighed Statistic	p-Value
Pedroni test
v	0.6546	0.2563	0.9626	0.1679
rho	−0.9851	0.1623	−1.1746	0.1201
PP	−2.6972	0.0035	−2.3859	0.0085
ADF	−3.2080	0.0007	−3.2468	0.0006
Grouped-rho	−0.3811	0.3516
Grouped-PP	−2.4324	0.0075
Grouped-ADF	−3.3281	0.0004
Kao test
Statistic	−2.5958	0.0047
Variance	0.0184
Fisher-Johansen by Maddala and Wu test
Cointegrating Vectors	Trace Statistic		Max-Eigen Statistic
0	44.24	0.0000	40.36	0.0001
1	14.47	0.2716	10.65	0.5592
2	22.14	0.0360	22.14	0.0360

and

Table 3. Cointegration in LCBC_it model.

	Statistic	p-Value	Weighed Statistic	p-Value
Pedroni test
v	0.7477	0.2273	0.5906	0.2774
rho	−0.9262	0.1772	−1.0806	0.1399
PP	−2.0042	0.0225	−1.7877	0.0369
ADF	−2.1793	0.0147	−2.0728	0.0191
Grouped-rho	−0.3088	0.3787
Grouped-PP	−1.7434	0.0406
Grouped-ADF	−1.6471	0.0498
Kao test
Stat	−1.0621	0.1441
Variance	0.0212
Fisher-Johansen by Maddala and Wu test
Cointegrating Vectors	Trace test		Max-Eigen test
0	23.79	0.0218	19.76	0.0717
1	12.13	0.4353	8.902	0.7113
2	20.16	0.0642	20.16	0.0642

show that Kao’s [73] test confirmed long-run association in the LTBC_it model but not in the LCBC_it model. On the other hand, Fisher-Johansen’s test found two cointegrating vectors in both LTBC_it and LCBC_it models. Further, Pedroni’s [75] test confirmed the cointegration in four out of seven statistics and two out of four weighted statistics. Hence, cointegration tests provided sufficient evidence of long-run relationships in the LCBC_it and LTBC_it models. Furthermore, coefficients of ECT_t-1 were negative in both LTBC_it and LCBC_it models in

Table 4. Regression results.

Variable	LTBCit				LCBCit
	Parameters	S.E.	t-Statistic	p-Value	Parameters	S.E.	t-Statistic	p-Value
FMOLS
LGEit	0.1281	0.0496	2.5848	0.0106	0.1664	0.0496	3.3584	0.0010
LMSit	−0.2091	0.0260	−8.0564	0.0000	−0.1765	0.0260	−6.8002	0.0000
DOLS
LGEit	0.0826	0.0849	0.9729	0.3326	0.2269	0.1204	1.8845	0.0620
LMSit	−0.3694	0.0743	−4.9737	0.0000	−0.2811	0.0727	−3.8664	0.0002
PMG
LGEit	0.2690	0.0823	3.2677	0.0013	0.3549	0.1308	2.7141	0.0074
LMSit	−0.4136	0.0783	−5.2789	0.0000	−0.3695	0.1217	−3.0371	0.0028
ECTt-1	−0.3512	0.1001	−3.5078	0.0006	−0.2584	0.0921	−2.8064	0.0057
ΔLTBCit-1	0.1264	0.0541	2.3373	0.0208
ΔLCBCit-1					0.0466	0.1203	0.3876	0.6988
ΔLGEit	0.3455	0.1127	3.0642	0.0026	0.1406	0.1050	1.3386	0.1828
ΔLMSit	0.2158	0.0973	2.2171	0.0281	0.3348	0.1449	2.3101	0.0223
Intercept	1.3669	0.3977	3.4371	0.0008	0.8291	0.3160	2.6242	0.0096
Pesaran cross-section dependence test	0.2184			0.8271	0.6853			0.4932

, which confirms the cointegration in the models. Hence, we may proceed with long- and short-run analyses.

Table 4 demonstrates the long-term positive impact of government spending on TBC_it and CBC_it in GCC countries in PMG, FMOLS, and DOLS results. The government expenditure is a direct component of aggregate demand, demonstrating a scale effect of fiscal policy in GCC economies in terms of both territory- and consumption-based CO₂ emissions. For example, increasing government spending increases energy demand and both TBC and CBC emissions, which shows that most of the energy demand is sourced from fossil fuel. The consumption-based emissions are adjusted by international trade. Our results show that fiscal policy discourages both production and consumption-based pollution-oriented economic activities. Another possible reason for this result is that oil is a major source of income and exports in GCC countries, which is pollution-oriented and accelerates CO₂ emissions in production and consumption activities. Our finding is also corroborated by literature investigating the impact of government aggregate expenditures on emissions [54,59]. However, the literature has also reported the positive environmental effect, through reducing emissions, of public spending in clean energy and green projects [47,48]. Moreover, some literature also found the negative long-term effect of aggregate public spending on emissions [40,48,49].

The effect of LMS_it was found to be negative in all estimates. Monetary policy is proxied by broad money, including cash outside the banks and all types of bank deposits, securities, and travelers’ cheques. Hence, money supply represents both liquid cash and financial sector deposits, which has a pleasant effect on the environment in GCC economies. This result corroborates that monetary policy in GCC countries is promoting cleaner technologies in the long run. This result is in line with Ullah’s et al. [40] findings, which elaborate that the monetary shocks need a long run to have positive environmental effects. Contrarily, the literature has found a positive impact of money supply on emissions [5,37,38].

In the short run, parameters of ECT_t-1 are negative and show divergence to a long-term equilibrium. The lagged term ΔLTBC_it-1 shows that TBC in any year is promoting TBC in the subsequent year. However, ΔLCBC_it-1 does not have any effect on subsequent years’ CBC emissions. Government spending in promoting TBC emissions does not affect CBC emissions. Past literature also verified our finding of the positive short-run effect of public expenditure on TBC emissions [40,42]. Moreover, Yuelan et al. [54] reported the opposite finding of the negative short-term effect of public spending on emissions. Monetary policy positively impacts both CBC and TBC emissions. However, this effect was negative in the long term, demonstrating that monetary policy needs a long time to have a positive environmental impact by reducing emissions. Ullah et al. [40] also reported the same positive and negative environmental effects of monetary policy in the long and short terms.

5. Conclusions

Macroeconomic policies can have profound effects on the environment of any economy or region. We investigated the role of monetary and fiscal policies on the TBC and CBC emissions in six GCC countries. We found the long-term relationships in cointegration analyses of TBC and CBC emissions models. Moreover, we applied various panel techniques to confirm the robustness of the research findings. In the long run, government expenditures have positive effects on both TBC and CBC and have a scale effect on GCC economies through increasing CO₂ emissions. Theoretically, government expenditure is one of the components of aggregate demand. Hence, increasing aggregate demand through fiscal policy has environmental consequences in releasing CO₂ emissions because fossil fuel is a major chunk of GCC economies’ energy mix. Moreover, government spending is also majorly sourced from oil production and revenues in the GCC economies. The positive effect of fiscal policy on both TBC and CBC corroborates that fiscal policy has a uniform impact on TBC and CBC emissions in the GCC region. Government spending shows a positive effect on TBC emissions in the short run, but it has an insignificant impact on CBC emissions. Money supply has a negative long-term effect on both TBC and CBC emissions and has a positive short-term effect on both TBC and CBC emissions. Hence, monetary policy has short-term environmental consequences in terms of TBC and CBC emissions in GCC countries. However, in the long run, monetary policy is helpful for reducing both TBC and CBC emissions in the GCC region.

5.1. Policy Recommendations

Based on the results, we recommend that the GCC economies care about the environmental consequences of government spending by encouraging renewable energy consumption in government projects to save the environment from CO₂ emissions. The monetary policy would reduce CO₂ emissions as per the findings of this research. Hence, the monetary policy should be encouraged further to promote a clean environment. Additionally, expansionary monetary policy should be encouraged to provide loans for green projects to promote renewable and energy-efficient technologies.

5.2. Limitations and Future Directions

The present research utilized a limited period of 1990–2019 due to the unavailability of CBC emissions data before 1990. Moreover, the analysis was restricted to GCC countries. To improve the generalization of the results, future research may extend the time range by generating CBC emissions data for an extensive range of time periods and might consider utilizing a panel of Middle East and North Africa countries to promote the scope of the present research.