Examining the drivers of CO2 Emission: Evidence from Indonesia

CO2 emissions are a significant risk and threat not only to the economy, environment, and agriculture but also to public health. In this study, we examine empirical evidence of the impact of economic growth, renewable energy use, and agricultural sector improvement on reducing CO2 emissions and environmental degradation in Indonesia based on the Environmental Kuznets Curve (EKC) hypothesis. This study uses annual time series data from 1988 to 2020 by applying Dynamic Least Squares (DOLS) and Fully-Modified Least Squares (FMOLS) econometric estimation techniques. Based on empirical evidence, the EKC phenomenon in Indonesia cannot be validated and justified because it correlates with increasing economic growth rates as well as increasing CO2 emissions. In addition, technological innovation and renewable energy applied to Indonesia’s more modern agricultural sector have a significant impact on reducing carbon emissions and improving environmental quality. Furthermore, the consumption of renewable energy in Indonesia is limited to the manufacturing and service sectors, which means it does not contribute to carbon emissions.


1.Introduction
Indonesia pledged to cut carbon emissions and environmental damage by 29% to 41% by 2030 at the Paris Agreement meeting, in 2016.In ASEAN, Indonesia produces the most carbon emissions with significant annual increases, putting 2,000 islands at risk of disappearing because of the effects of changing the environment.[1].The Indonesian government is taking active action in mitigating carbon emissions because it is investing 200 USD billion/year over the nine years budgeted in the fiscal [2] Developing countries including Indonesia, which have natural resources and pursue economic growth, will damage the environment more quickly because they are still dependent on fossil energy.It is associated with the hypothesis of the Environmental Kuznets Curve (EKC) that proposes that carbon emissions and growth in the economy are inversely proportional, at the initial stage increases but when income increases carbon emissions return to the equilibrium point [3], [4].Previous researchers [5] and 1302 (2024) 012070 IOP Publishing doi:10.1088/1755-1315/1302/1/012070 2 [6] validated the EKC hypothesis.In contrast, [5] and [7] is found not supported the EKC in the United States, Vietnam, and West Africa failed to identify the EKC hypothesis.
[8], Extensive economic growth in OECD countries for consumption, production, and transportation is highly dependent on fossil energy sources which produce CO2 emissions.[9], that significant economic growth correlates with increasing CO2 emissions in India because of using energy from fossil sources.In addition, [10] states the American economy will always grow without destroying the environment and pollution due to the launch of renewable technology and energy.[11] stated that there is a government policy to fulfils renewable energy-RE sources using a combination of biodiesel and fossil fuels to encourage demand for palm oil internationally.[12] Inclusive and zero carbon economic growth is always driven by investment in infrastructure, government policies, and technological advances to drastically reduce the intensity of CO2 emissions.
Indonesia's agricultural sector is the main sector contributing to absorbing 27.73% of the workforce and contributing 13.7% of GDP [13].Conversely, in the field of agriculture.is very vulnerable to climate change, sea level rise, and salinity [14].Indonesia's agricultural sector contributes 14% of global carbon emissions, but by 2050 it is estimated to increase by 30% due to land use changes, deforestation, urbanization, and climate change [15].Increasing agricultural production capacity using conventional systems and the use of fossil fuels raises carbon emissions negatively [16].
In the study of [17] changes in agricultural land use will increase CO2 emissions.The increased use of fossil energy and combustion in the agricultural sector has contributed significantly to environmental deterioration and rising CO2 emissions in Peru [18].[19] said government requires the use of CO2 to increase the sustainability of its development.[16] also found that rapid urbanization with deforestation significantly increased CO2 emissions and climate change in Malaysia.Renewable energy in the agricultural sector increases agricultural production and significantly reduces short-and long-term CO2 emissions in North Africa [20].Renewable energy and innovative technology raise the standard of the environment by providing benefits to increased agricultural production in lower, middle-and highincome countries [21].
Recently, renewable energy was integrated to reduce environmental damage while maintaining economic development.The increasing price of energy derived because of using energy from fossil sources and the shortage of energy that is not renewable supplies make it increasingly important for countries to immediately switch to adopting renewable energy [18].[22,23,24] RE reduces CO2 and doesn't impede to expansion of economies in both developed and emerging nations.[25] found that investment in renewable energy resources will increase GDP and create new jobs without damaging the environment.
The environment and renewable energy have a link that has become an interesting issue because it forces countries to reduce dependence on fossil energy [26].India is turning to renewable energy which significantly reduces carbon emissions [27].[28] find that in ASEAN increased Both carbon emissions and reliance on non-renewable energy are decreased by renewable energy.[29], discovered that emissions of carbon dioxide (renewable energy) have impact positively and negatively impacted by energies that are not renewable in Mediterranean region over the period 1980-2014.
This study wants to explore the potential for reducing emissions in Indonesia using an econometric approach, which is a hot topic among academics throughout the world.To fill existing knowledge gaps, The purpose of this study is to investigate the dynamic effects of development in the economy, agricultural innovation, and sources of RE on Indonesia's carbon dioxide emissions using DOLS, and FMOLS methods.This study is interesting to do because Indonesia ranks 1st in ASEAN and 16th globally in terms of GDP of 3328 USD [14].On the other hand, Indonesia has vast agricultural land and abundant energy resources, so the state's role is important in designing policy mechanisms and the application of energy from renewable sources.

Data
This study makes use of time series data from 1988-2020 obtained from www.data.worldbank.Where CO2 emissions kt [CO2] is the dependent variable, and GDP per capita growth annual [GDP], agricultural growth (Agriculture, forestry, fishing, and value-added of GDP) [ARG], and is utilization of the overall final energy used [RE] are independent variables.We standardized all variables to the natural logarithm to address heteroscedasticity.

Research model
In this study, we use both methods modified and non-modified Ordinary Least Squares (DOLS, and FMOLS) to estimate the model.The advantage of DOLS is that it can resolve cointegration, autocorrelation, and endogeneity bias as well as integrate more specific aspects of lead and lag in control to produce valid standard deviation covariance matrices and significance between variables.Meanwhile, FMOLS uses a non-parametric approach and is suitable for small samples because it produces consistently valid estimates and is efficient.Before carrying out estimates, a stationarity test, classical assumption test, and CUSUM test are first carried out to see the stability of the research model.

Result and discussion
Table 1 presents the mean, minimum, standard deviation, maximum, and descriptive statistics of each variable, and J-Bera are shown in panel A. There are 22 time series data from Indonesia for every variable.The Jarque-Bera statistical value for each variable is greater than 5% concluding that the validity of the data is normal in this study.Panel B shows the stationarity results using Augmented-Dickey-Fuller (ADF) and Phillips-Perron (PP) tests presented as prerequisites for estimating the DOLS and FMOLS models.In this investigation, all variables are first-order stationary rather than stationary at the level different so that the resulting estimates are unbiased and valid.Panel C passes the classical assumption test.Panel C passes the classical assumption test.In addition, to evaluate and see the long-term stability of the estimated model, this research uses the cumulative sum of recursive residuals (CUSUM) and cumulative sum of squares of recursive residuals (CUSUMQ) test plots (Figure 1).The results of Figure 1 above show that it is significant at the 5% level in both the CUSUM and CUSUMQ plot tests.In the graph, the red line shows the confidence interval of the model estimate, while the residual is shown on the blue line.The test results of this research model are valid because the 5% residual significance is in the 5% interval so there is no change in the long run.based on the results of stationarity testing, assumption tests, and long-term capital stability, the DOLS and FMOLS model estimates are met.
Table 2 shows the long-run estimation results of DOLS and FMOLS.Both estimation results provide similar and consistent results in terms of coefficients and statistical significance.The trend rate of a 1% increase in the LNGDP variable in Indonesia, causes a significant increase at the 0.5% level which increases CO2 by 0.8288 and 0.8099.This theme apparently shows that the EKC hypothesis in Indonesia using this research data is that this theme is not fulfilled or validated.Indonesia's economic growth has led to a decrease in environmental degradation and pollution resulting in increased CO2 emissions in the long term.This is due to Indonesia's lack of renewable energy production, expensive renewable energy costs and lack of technological innovation so that in producing manufacturing and services there is still a very large use of fossil energy.These results are like those found by [16]; and [6] Economic improvement is associated with increased public consumption which causes excessive use of fossil energy, pollution, and waste.
Based on the statistical results, in Indonesia, it turns out that in the long run, an increase in the LnARG variable has a negative and significant impact at the 5% level on CO2 emissions, which means that an increase in agricultural production by 1% will result in a decrease in CO2 emissions and improve environmental quality by 0.2670% and 0.2872%.This finding is also relevant to the research conducted by [21]; and [30], This shows that this sector has started to switch to environmentally friendly energy, environmentally friendly fertilizers, and adoption of technology that contributes to improving environmental quality and reducing CO2 emissions.The process of transition to renewable energy and technology can facilitate the process of integration of production, is cost-effective and clean, thereby significantly reducing environmental degradation and helping to increase a country's economic growth [31].
Moreover, in the long run, the coefficient of the LnRE variable is negative but insignificant at the 10% level, implying that renewable energy does not have a significant impact on CO2 emissions in Indonesia.Although this result is different from the findings of [27] and [28], it is consistent with the findings of [32], even though Indonesia has enormous renewable energy potential.
Indonesia's renewable energy use is not an important factor in determining the sustainability of economic growth and CO2 emission reduction.The use of fossil energy mix in Indonesia is still greater than renewable energy in the production of manufacturing and services so CO2 emissions have not decreased significantly because of the use of renewable energy.[33] The percentage contribution of the renewable energy mix was only 14.71% in 2020 due to stagnant technological development and innovation due to a lack of investment in renewable energy resources.

Conclusions
Empirical findings show that environmental degradation and pollution are increasing along with increasing economic growth.This shows strong evidence against the EKC hypothesis.Second, growing agriculture reduces carbon emissions and enhances the environmental quality due to the adoption of technology and RE.Third, RE is not impacted for greenhouse gas emissions, this shows that its use is still low and inefficient.Policymakers in Indonesia must take appropriate steps to reduce carbon emissions without reducing economic factors.In addition, future research can add ASEAN countries and institutional quality moderating variables.

Figure 1 .
Figure 1.The result of the cumulative sum of recursive residuals (CUSUM) and the sum of squares of recursive residuals (CUSUMQ).