INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4948 www.rsisinternational.org
Malaysia’s Renewable Energy and CO₂ Emissions: ARDL–ECM
Evidence with Policy and Managerial Implications
Azrizal Husin1*, Nur Hayati Abd Rahman2, Mohd Danial Afiq Khamar Tazilah3, Heri Yanto4, Niswah
Baroroh4
1*Faculty of Management and Economics, Universiti Pendidikan Sultan Idris, 35900 Tanjung Malim
Perak, Malaysia
2Faculty of Business and Management, Universiti Teknologi MARA Kampus Alor Gajah, 78000 Alor
Gajah Melaka, Malaysia
3Faculty of Management and Economics, Universiti Pendidikan Sultan Idris, 35900 Tanjung Malim
Perak, Malaysia
4Faculty of Economics and Business, Universitas Negeri Semarang, Semarang, Indonesia
*Corresponding Author
DOI: https://dx.doi.org/10.47772/IJRISS.2025.910000407
Received: 12 October 2025; Accepted: 19 October 2025; Published: 13 November 2025
ABSTRACT
This study investigates whether renewable energy (RE) adoption reduces carbon dioxide (CO₂) emissions in
Malaysia and under what conditions the effect is durable. We focus on translating empirical results into decisions
for Malaysian firms and policymakers, given continuing reliance on coal-fired power and accelerating—but
uneven—RE deployment. A quantitative, time-series design is employed using annual national data. An
autoregressive distributed lag (ARDL) model and its error-correction representation (ECM) are used to separate
short-run dynamics from long-run equilibrium relationships. The dependent variable is CO₂ emissions, with RE,
GDP growth (economic scale/EKC context), foreign direct investment (FDI), and forest cover as regressors.
Model diagnostics address integration properties, lag selection, and stability. Results show that RE has a negative
and statistically significant short-run effect on CO₂—consistent with dispatch substitution in which incremental
RE output crowds out marginal coal/gas generation. Long-run elasticities for RE and forest cover are
directionally negative but statistically imprecise in the baseline specification, while GDP remains emissions-
increasing. The ECM coefficient is negative and significant, indicating relatively fast adjustment toward long-
run equilibrium. These findings imply that Malaysia can realise near-term abatement via RE procurement and
efficiency measures, but durable decarbonisation requires policy sequences that convert substitution into
structural displacement—specifically, competitive RE auctions paired with storage and scheduled coal
retirements, complemented by green-halo FDI and credible forest governance. The study contributes Malaysia-
specific, horizon-separated evidence that informs both managerial energy procurement choices and the state’s
transition planning.
Keywords: Malaysia, Renewable Energy, ARDL, ECM, CO₂ Emissions, EKC, FDI, Forest Cover, Energy
Transition.
INTRODUCTION
Global climate change has become one of the most urgent challenges of our time, with carbon dioxide (CO₂)
emissions recognized as the primary driver of global warming and environmental degradation. As countries
strive to balance economic development with environmental sustainability, the role of renewable energy has
gained increasing attention. Renewable energy sources such as solar, hydro, biomass, and wind offer cleaner
alternatives to fossil fuels and represent a crucial pathway for reducing carbon intensity in national economies.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4949 www.rsisinternational.org
Malaysia, as an upper-middle-income nation, is highly dependent on fossil fuels, particularly coal and natural
gas, to meet its energy demand. This reliance has contributed to a consistent rise in CO₂ emissions over the past
decades. Projections indicate that emissions are likely to continue increasing beyond 2023, largely due to the
dominance of coal-fired power generation, unless significant changes are made. The high emission intensity of
Malaysia’s energy sector reflects the structural challenges of transitioning to a low-carbon economy.
In recognition of this challenge, Malaysia has pledged under its Nationally Determined Contribution (NDC) to
reduce greenhouse gas (GHG) intensity of GDP by 45% by 2030 compared to 2005 levels (Vaicondam et al.,
2024), and to achieve net-zero emissions by 2050 (Aminuddin et al., 2025). In order to operationalize this
commitment, the government launched the Malaysia Renewable Energy Roadmap (MyRER 2022–2040) and
the National Energy Transition Roadmap (NETR 2023). These policies aim to increase renewable energy
capacity from only 4% in 2023 to 31% by 2025 and 40% by 2035, alongside a gradual phase-out of coal.
Concrete measures have already been undertaken to support this transition. For example, the government has
targeted the transportation sector, one of the largest contributors to national emissions, by promoting electric
vehicles (EVs), introducing incentives for EV adoption, and investing in charging infrastructure (Songkin et al.,
2023). Public transportation systems have also been expanded and modernized to reduce dependence on private
vehicles (Shahid et al., 2014).
Beyond energy and transport, the government has also acted on the agriculture and plantation sector, particularly
palm oil, which is both vital to Malaysia’s economy and a source of significant environmental impact. The
government has implemented policies to promote sustainable palm oil production and to reduce emissions from
this sector, including the adoption of the Malaysian Sustainable Palm Oil (MSPO) certification scheme
(Jamaludin et al., 2024). These initiatives aim to balance economic reliance on palm oil exports with Malaysia’s
environmental commitments, ensuring that growth in this sector aligns with the broader national strategy for
carbon reduction.
Given these developments, the relationship between renewable energy adoption and CO₂ emissions in Malaysia
warrants careful examination. While renewable energy is theoretically linked to lower emissions, the pace, scale,
and effectiveness of its adoption in Malaysia remain critical questions. Assessing whether renewable energy
expansion is sufficient to curb the rising trend of CO₂ emissions is central to understanding Malaysia’s ability to
meet its climate commitments and transition to a sustainable energy future.
Problem Statement
Although Malaysia has acknowledged the urgency of reducing CO₂ emissions and introduced ambitious
roadmaps, emissions are still on an upward trend albeit with a gradual decline, surged from 28 million tonnes
(Mt) in 1980 to 262.2 Mt in 2020 (Segar et al., 2024), indicating a ninefold increase. The energy sector remains
heavily reliant on coal, which is the most carbon-intensive fuel source, and renewable energy adoption, while
expanding, still represents a relatively small share of the overall energy mix.
Existing research on Malaysia’s energy transition often highlights the technical potential of renewable energy
but provides limited empirical evidence on its actual effect in reducing CO₂ emissions. Moreover, projections
suggest (Segar et al., 2024) that unless renewable energy adoption accelerates significantly, Malaysia’s
emissions will continue to rise in the coming decade despite policy targets.
This creates a pressing research gap: it is still unclear to what extent renewable energy adoption in Malaysia can
effectively mitigate CO₂ emissions in line with national and international climate goals. Therefore, this study
seeks to address the following core question:
Does renewable energy (RE) adoption significantly reduce CO₂ emissions in Malaysia?
Answering this question is vital for policymakers and stakeholders to evaluate the effectiveness of existing
strategies and to design stronger interventions that align Malaysia’s energy transition with its emission reduction
targets.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4950 www.rsisinternational.org
LITERATURE REVIEW
Malaysia’s energy–emissions context and conceptual lenses
Malaysia’s carbon outcomes reflect the interaction of its electricity fuel mix, macroeconomic expansion, inward
capital, and land-use governance. Coal’s rising share has historically locked in emissions intensity, while
renewable uptake—dominated by hydro with growing but still modest non-hydro contributions—remains
insufficient to reliably baseload fossil generation at scale (Lau, Tan, & Tang, 2016; Lau et al., 2018). Within
standard production–energy frameworks, renewable energy can reduce emissions when it displaces fossil inputs
rather than simply meeting incremental demand; foreign direct investment may raise emissions via a pollution-
haven channel or reduce them through a green-halo mechanism that transfers cleaner technologies and
managerial practices (Naz et al., 2019; Pham, Nguyen, Nguyen, & Tran, 2025; Zhao, 2025). Economic growth
typically raises emissions at early development stages (scale effect), with any Environmental Kuznets Curve
flattening contingent on structural efficiency and energy-mix change (Saboori, Sulaiman, & Mohd, 2016; Sarkar,
Al-Amin, Mustapa, & Ahsan, 2019). Forested area acts as a carbon sink with effects that emerge primarily over
longer horizons (Begum, Raihan, & Said, 2020; Ponce, Del Río-Rama, Álvarez-García, & Oliveira, 2021; Suman
et al., 2025). Malaysia-focused and regional/cross-country studies in the evidence base also provide design cues
(ARDL/NARDL; threshold VAR) for interpreting short-run versus long-run dynamics (Akhtar, Masud, Al-
Mamun, & Saif, 2023; Zhao, 2025).
Renewable energy → CO₂: what differs in the short run versus the long run
Short run. For Malaysia, the most immediate carbon relief from renewable energy arises via dispatch
substitution—incremental hydro or other RE output crowding out marginal coal or gas generation within the
year. Malaysia-specific evidence documents dynamic linkages among hydro use, growth, and emissions that are
indicative of near-term substitution rather than structural retirement (Lau, Tan, & Tang, 2016).
Complementarily, Malaysian studies show that RE consumption is shaped by macro drivers, implying that short-
horizon RE shocks chiefly operate through substitution channels (Lau et al., 2018). Regional and multi-country
designs reinforce the short-run mitigation result but flag sensitivity to measurement (RE electricity share vs.
total final energy) and to displacement (whether RE truly replaces fossil generation): threshold-VAR evidence
shows robust mitigation from RE while highlighting regime dependence that can shape the net carbon effect
when macro conditions vary (Zhao, 2025).
Long run. Over longer horizons, studies generally find that higher RE penetration reduces CO₂, but statistical
precision hinges on (i) whether additions displace coal rather than merely accommodate demand growth, (ii) the
metric used for RE, and (iii) broader innovation capacity. Cross-national evidence associates higher RE and
R&D human capital with lower emissions, and shows that RE’s long-run mitigation persists across macro
regimes, although magnitudes vary with thresholds (Mentel et al., 2022; Zhao, 2025). For Malaysia, prior work
suggests that deeper and more diversified RE penetration is needed to translate short-run substitution into precise
long-run displacement effects, a view consistent with the country’s still-modest non-hydro scale (Lau et al.,
2018).
Economic growth and the Environmental Kuznets Curve (EKC) debate in Malaysia
On average, Malaysian studies report that economic growth raises CO₂ in the short run, reflecting scale effects
that dominate before efficiency or energy-mix changes materialize. Saboori, Sulaiman, & Mohd (2016) and
Sarkar, Al-Amin, Mustapa, & Ahsan (2019) both find cointegrated GDP–CO₂ relationships consistent with
positive elasticities, with any EKC-type turning point dependent on policy-driven structural shifts rather than
automatic with income growth. Regional comparators (e.g., Brunei) report similar growth–emissions links where
decoupling follows explicit energy-system reforms. (e.g., Xuan, 2025).
FDI → CO₂: pollution haven, green halo, and thresholds
Malaysia’s FDI–emissions nexus is conditional on sectoral composition, absorptive capacity, and macro
regimes. Using a NARDL design, Akhtar, Masud, Al-Mamun, & Saif (2023) document emissions linkages to
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4951 www.rsisinternational.org
FDI alongside energy and growth, consistent with exposure to pollution-haven dynamics when inflows favor
energy-intensive manufacturing. A Malaysia-focused study on ecological pressures by Chuah, Cheam, Chua, &
Arshad (2025) likewise shows sensitivity to composition and regime, indicating that the sign of FDI’s impact
cannot be assumed ex ante. In a 27-country panel threshold-VAR, Zhao (2025) finds that FDI’s net effect can
flip under higher growth thresholds, while renewable energy retains a mitigating role—evidence that macro
conditions shape both the size and direction of FDI’s carbon impact. Robust least squares estimate in Naz et al.
(2019) further corroborate the importance of renewable penetration and growth regimes when assessing the FDI–
emissions channel. At the same time, cross-country work detects a green-halo under stronger absorptive capacity
and innovation linkages: Pham, Nguyen, Nguyen, & Tran (2025) show that FDI can co-move with lower
emissions when complementary conditions are present, consistent with the view that sectoral screening and
targeting condition FDI’s environmental effects.
Forested area → CO₂: sequestration over time
Forested area increases generally reduce CO₂ via carbon sequestration, but effects materialize over longer
horizons. For Malaysia, Begum, Raihan, & Said (2020) report dynamic impacts of growth and forested area on
emissions consistent with a carbon-reducing role of forests. In broader contexts, ARDL evidence links forest
conservation and renewable energy to lower emissions (Ponce, Del Río-Rama, Álvarez-García, & Oliveira,
2021). In G20 settings, improvements in clean energy access and forest cover are likewise associated with lower
emissions, underscoring the importance of sustained land-use policy alongside energy transitions (Suman et al.,
2025).
In the short-run forest effects are often small or imprecise, consistent with slow biophysical sequestration and
gradual land-use adjustment; consequently, many studies detect forest impacts primarily in the long run (Begum
et al., 2020; Ponce et al., 2021; Suman et al., 2025).
Synthesis and gap addressed by this study
Three robust takeaways emerge from the evidence base. First, renewables mitigate CO₂ in short-run horizons
through substitution—particularly where hydro or fast-dispatch RE is available—whereas long-run precision
depends on whether additions displace coal and on how renewable energy is measured (Lau, Tan, & Tang, 2016;
Lau et al., 2018; Zhao, 2025; Mentel et al., 2022). Second, in Malaysia, economic growth remains emissions-
increasing unless accompanied by structural energy changes and efficiency gains; EKC-type turning points are
not automatic at Malaysia’s current income level (Saboori, Sulaiman, & Mohd, 2016; Sarkar, Al-Amin, Mustapa,
& Ahsan, 2019). Third, the environmental effect of FDI is context-dependent: pollution-haven channels tend to
dominate when inflows favor energy-intensive sectors, whereas green-halo effects emerge under stronger
absorptive capacity and favorable macro regimes (Akhtar, Masud, Al-Mamun, & Saif, 2023; Naz et al., 2019;
Pham, Nguyen, Nguyen, & Tran, 2025; Zhao, 2025). In addition, forest coverage reduces emissions over time,
while short-run forest coefficients are often small or imprecise, consistent with sequestration dynamics and
gradual land-use adjustment (Begum, Raihan, & Said, 2020; Ponce, Del Río-Rama, Álvarez-García, & Oliveira,
2021; Suman et al., 2025).
Bridge to the empirical design (gap statement). Prior work in Malaysia rarely spans a broad period covering
major policy shifts while using a composite renewable metric and jointly accounting for GDP growth, FDI, and
forested area within an ARDL/ECM that reports both short-run and long-run elasticities. To address this gap,
the present study estimates such a specification to test whether Malaysia’s current renewable penetration is
associated with short-run substitution and long-run displacement effects under local macro and sectoral
conditions.
METHODOLOGY
This study employs annual time-series data for Malaysia covering the period 1991–2024. The dataset was
obtained from the World Development Indicators (WDI) published by the World Bank. The variables used are
as follows:
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4952 www.rsisinternational.org
Table 1: Variables Used in This Research
No. Notation Type Indicator Name
1 lco2 Dependent variable Carbon dioxide emissions as a percentage of total emissions
2 lrenew Independent variable Renewable energy consumption as a percentage of total final
energy consumption
3 lgrowth Control variable GDP (annual % growth)
4 lforest Control variable Forest area (% of land area)
5 lfdi Control variable Foreign direct investment, net inflows (% of GDP)
All variables were transformed into natural logarithmic form to reduce potential heteroskedasticity, improve
distributional properties, and allow elasticity-based interpretations of coefficients.
The relationship between renewable energy consumption and carbon dioxide emissions was estimated using the
Autoregressive Distributed Lag (ARDL) approach, developed by Pesaran and Shin (1999). The ARDL
methodology is particularly suitable for small sample sizes and for data that are integrated of mixed orders, i.e.,
I(0) or I(1).
The general ARDL model can be specified as:
lco2t =∝0+∑∅i
p
i=1
lco2t−1 +∑βj
q1
j=0
lrenewt−j +∑γk
q2
k=0
lgrowtht−k + ∑ δm
q3
m=0
lforestt−m +∑θn
q4
n=0
lfdit−n + εt
[Equation 1]
Before estimating the ARDL model, it is necessary to determine the order of integration of each series.
Conducting unit root tests such as the Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) is crucial to
avoid spurious regression results. If the variables are found to be I(2), the ARDL approach is no longer valid.
Besides, the correct lag length is essential to capture the underlying dynamics of the model. Using too few lags
can lead to omitted variable bias, while too many lags reduce efficiency and may create multicollinearity.
Therefore, lag length selection criteria such as the Akaike Information Criterion (AIC), Hannan-Quinn (HQIC),
Schwarz Bayesian Criterion (SBIC), and Final Prediction Error (FPE) were employed.
The ARDL model can be re-parameterized into an Error Correction Model (ECM) form that incorporates both
short-run dynamics and the long-run equilibrium adjustment mechanism:
∆lco2t =∝0+∑∅i
p−1
i=1
∆lco2t−1 + ∑ βj
q1−1
j=0
∆lrenewt−j + ∑ γk
q2−1
k=0
∆lgrowtht−k + ∑ δm
q3−1
m=0
∆lforestt−m
+ ∑ θn
q4−1
n=0
∆lfdit−n +φECMt−1 + εt
[Equation 2]
where:
ECMt−1 is the lagged error correction term capturing deviations from long-run equilibrium,
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4953 www.rsisinternational.org
φ is the speed of adjustment coefficient.
A negative and statistically significant value of φ confirms convergence toward long-run equilibrium. The ECM
coefficient is derived from the lagged dependent variable of the ARDL model. Specifically:
φ = 1 −∑∅i
[Equation 3]
where ∅i are the coefficients of the lagged dependent variables in levels.
ARDL/ECM accommodates mixed integration orders commonly observed in annual macro–energy series and
yields short-run and long-run elasticities. Following best practice, we note that measurement choices for RE
(electricity share vs total final energy) and CO₂ (per-capita vs intensity vs total) can shift elasticity magnitudes;
therefore, we report diagnostics and interpret estimates with these differences in mind.
FINDINGS
Unit root tests
The stationarity properties of the series were examined using the Augmented Dickey-Fuller (ADF) and Phillips-
Perron (PP) tests. The results are illustrated in Table 2 below.
Table 2: Unit Root Test Results
Variable ADF Statistic PP Statistic Integration Order
lco2 -4.666*** -7.860*** I(0)
lrenew -1.205 -5.857*** I(1)
lgrowth -1.922 -1.295 I(1)
lforest -4.289*** -2.259 I(0)
lfdi -3.361 -5.401*** Mixed (I(0)/I(1))
Notes: Null = unit root. *** p < 0.01.
Results indicate that variables are integrated of mixed orders, I(0) and I(1), thereby validating the use of the
ARDL methodology.
ARDL estimation results
The optimal lag structure was determined using information criteria. The ARDL (2,0,2,0,1) specification was
selected based on the Akaike Information Criterion (AIC) and Final Prediction Error (FPE).
The ARDL (2,0,2,0,1) results are reported below in Table 3, with R2 and adjusted R2 values of 0.733 and 0.607,
respectively.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4954 www.rsisinternational.org
Table 3: Results of ARDL (2,0,2,0,1).
Variable Coefficient Std. Error t-Statistic p-Value
Long-run relationship
lgrowth -1.421 3.903 -0.36 0.720
lrenew -1.883 5.051 -0.37 0.713
lforest -106.291 208.254 -0.51 0.616
lfdi -3.252 8.577 -0.38 0.709
Short-run dynamics
LD.lco2 -0.658*** 0.199 -3.31 0.004
D1.lrenew -0.513** 0.215 -2.39 0.028
LD.lrenew -0.335 0.200 -1.68 0.110
D1.lfdi -0.047* 0.025 -1.85 0.080
Constant -12.582 11.697 -1.08 0.296
ECM(-1) -0.972*** 0.214 -4.54 0.000
Notes: ECM(-1) is the error correction term. *** p < 0.01, ** p < 0.05, * p < 0.10.
Based on the above results, the long-run coefficients show negative associations between renewable energy,
GDP growth, forest area, FDI inflows, and carbon emissions, although not statistically significant.
The directionally negative yet statistically insignificant long-run elasticities for renewables, GDP, forest area,
and FDI are plausible in Malaysia’s context of coal lock-in and measurement sensitivity. Demand growth and
grid constraints can further dilute steady-state mitigation unless renewable penetration scales sufficiently to
displace coal rather than merely meet incremental load (Lau et al., 2018; Zhao, 2025). Malaysia’s hydro-centric
dynamics suggest that renewable additions have not yet reached a scale that displaces baseload fossil generation,
which helps explain imprecise steady-state effects (Lau, Tan, & Tang, 2016). The macro-driven nature of
renewable consumption further supports the view that recent gains reflect aggregate conditions rather than
structural retirement of fossil capacity (Lau et al., 2018). Regarding FDI, panel evidence indicates regime
dependence—its net environmental effect can vary across growth thresholds—and green-halo benefits tend to
materialize only under stronger absorptive capacity (Naz et al., 2019; Zhao, 2025; Pham, Nguyen, Nguyen, &
Tran, 2025). The lack of long-run precision for forest area is consistent with slow biophysical sequestration and
land-use adjustment dynamics documented in Malaysia and broader samples (Begum, Raihan, & Said, 2020;
Ponce, Del Río-Rama, Álvarez-García, & Oliveira, 2021).
Meanwhile, in the short run, renewable energy consumption significantly reduces carbon emissions, while FDI
inflows exert a modest negative effect at the 10% level.
The significant short-run reduction in CO₂ associated with higher renewable consumption is consistent with
dispatch substitution, whereby incremental renewables (notably hydro or fast-dispatch resources) crowd out
marginal coal or gas generation (Lau, Tan, & Tang, 2016; Lau et al., 2018). The small, negative short-run FDI
coefficient aligns with a nascent green-halo under favorable macro regimes, as observed in Malaysia-focused
and panel designs (Chuah, Cheam, Chua, & Arshad, 2025; Zhao, 2025).
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4955 www.rsisinternational.org
Converting short-run substitution into durable long-run decarbonization requires (i) coal-to-RE displacement
rather than capacity addition, (ii) grid flexibility investments (storage, interconnections, system services), and
(iii) FDI screening/targeting toward low-carbon, technology-transferring sectors, consistent with regime-
dependence and absorptive-capacity findings above.
The ECM (-1) value of -0.972 is negative and highly significant. This indicates that 97% of disequilibrium in
carbon emissions is corrected within one year, suggesting a very fast adjustment back to the long-run equilibrium
following short-term shocks.
Diagnostic Tests
Diagnostic tests were conducted to ensure model validity. The Breusch-Godfrey LM test was used to detect
serial correlation, while the Breusch-Pagan/Cook-Weisberg test examined heteroskedasticity.
Table 5: Diagnostic Tests
Test Statistic p-Value Conclusion
Breusch-Godfrey LM (Autocorrelation) H0 = No serial
correlation
χ² (1) = 0.256 0.613 No serial correlation
Breusch-Pagan / Cook-Weisberg (Heteroskedasticity)
H0 = No heteroskedasticity
χ² (1) = 3.08 0.079 No heteroskedasticity
(10% level)
Both tests confirmed the absence of econometric problems, indicating that the estimated ARDL model is robust
and reliable.
CONCLUSION
This paper set out to answer a simple but consequential question: does Malaysia’s push toward renewable energy
actually lower carbon emissions—and on what timeline? The issue matters because Malaysia’s power system
has long relied on coal and gas while electricity demand keeps rising. Policymakers have announced ambitious
transition plans, but if new renewables merely meet incremental load rather than replace fossil baseload, the
country could add capacity without cutting emissions. Understanding whether renewable energy delivers short-
run relief, long-run displacement, or both is therefore central to credible pathways for meeting national climate
goals while safeguarding growth and energy security.
Our results speak in plain terms. In the short run, more renewable energy is associated with lower emissions,
consistent with dispatch substitution—renewables crowd out the marginal coal or gas unit when they are
available. Foreign direct investment, on balance, shows a small near-term improvement in environmental
performance, suggesting that recent inflows may carry cleaner processes or technologies. In the long run, the
elasticities for renewables, growth, forests, and FDI are negative in direction but not estimated precisely. This
pattern fits a system in transition: renewables are helping at the margin, yet scale, diversification, and grid
readiness have not reached the point where fossil baseload is reliably displaced; forest sequestration works, but
slowly; and FDI’s effect remains contingent on sectoral composition and macro conditions.
Theoretically, the study clarifies when and how the energy–emissions nexus operates in a middle-income, coal-
reliant system. First, it distinguishes substitution from displacement: renewable additions can reduce emissions
on impact without guaranteeing precise long-run mitigation unless they replace baseload generation. Second, it
underscores regime dependence in the capital–environment channel: the sign and strength of FDI’s effect vary
with growth phases and absorptive capacity, reconciling mixed findings in the literature. Third, it links
biophysical lags to empirical detectability: forest cover contributes to mitigation over extended horizons, which
explains why short-run coefficients often look weak even when the long-run contribution is real. Together, these
insights integrate energy systems thinking with macro-environmental econometrics, offering a framework for
interpreting seemingly inconsistent short-run and long-run results.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4956 www.rsisinternational.org
For practice and policy in Malaysia, the message is to convert short-run substitution into durable long-run
displacement. The Energy Commission, the Single Buyer, and the system operator should align capacity
procurement and dispatch with explicit coal down-ramp schedules so new renewable megawatts are tied to
measurable fossil-generation reductions, not just headline capacity targets. Tenaga Nasional Berhad and planners
should accelerate grid-flexibility investments—storage, fast-ramping reserves, enhanced ancillary-services
markets, and strategic interconnections—so variable renewables can reliably replace baseload output. The
Economic Planning Unit and investment agencies can tilt incentives toward green FDI—grid technologies,
storage, efficiency services—using performance-based criteria that favor demonstrable emissions improvements
and technology transfer. State forestry departments and federal agencies should pursue long-horizon forest
strategies (avoided deforestation, restoration with robust monitoring) recognizing that benefits accumulate over
years, not quarters. Central ministries can improve measurement by publishing emissions-based outcome metrics
(coal MWh reduced, tonnes of CO₂ avoided) alongside capacity figures, and by reporting renewables both as
electricity share and total final energy to keep policy focused on outcomes rather than inputs.
These contributions come with limitations. The analysis relies on national aggregates for emissions and energy,
which cannot capture sectoral or plant-level heterogeneity. Variable definitions—such as using a composite
renewable share and an aggregate CO₂ metric—shape elasticities and may mask technology-specific dynamics
(for example, hydro versus non-hydro renewables). Although the ARDL framework addresses mixed orders of
integration and provides a transparent short-run/long-run decomposition, it cannot fully resolve endogeneity
concerns or pin down causal mechanisms under all regime changes and structural breaks. Finally, the time-series
perspective prioritizes national dynamics; it does not benchmark Malaysia’s progress against ASEAN peers
facing similar constraints.
These limits point directly to the next research steps. Future work should (i) open the black box of FDI by
disaggregating inflows by sector to separate pollution-haven from green-halo channels and to test for threshold
effects explicitly; (ii) refine measurement by using technology-specific renewable series (hydro, solar, wind,
biomass) and alternative emissions metrics (per-capita, intensity, sectoral) to assess robustness; (iii) probe
regime dependence with asymmetric and threshold ARDL designs and structural-break tests aligned with major
policy changes, fuel-price shocks, or drought years; (iv) strengthen identification using plausibly exogenous
instruments—such as hydrology-driven variation or global fuel-price shocks—to sharpen the interpretation of
short-run substitution versus long-run displacement; and (v) benchmark regionally with comparable ASEAN
panels to place Malaysia’s displacement progress and policy effectiveness in context. Advancing along these
lines will move the debate from whether renewables help to how to design markets, grids, and investment policy
so near-term gains harden into sustained, economy-wide decarbonization.
ACKNOWLEDGEMENT
This research is funded by GERAN KOMPREHENSIF UNIVERSITI-GPKU, Universiti Pendidikan Sultan Idris
(UPSI), Perak, Malaysia, Grant Number (2025-0026-106-01).
REFERENCES
1. Akhtar, R., Masud, M. M., & Al-Mamun, A., & Saif, A. N. M. (2023). Energy consumption, CO₂
emissions, foreign direct investment, and economic growth in Malaysia: An NARDL technique.
Environmental Science and Pollution Research. https://doi.org/10.1007/s11356-023-26246-4
2. Aminuddin, F. A., Baihakki, A. S., & Kamaruddin, T. (2025). Roles of local authorities in carbon
emission reduction: Case studies from three municipalities in Johor Bahru. IOP Conference Series: Earth
and Environmental Science, 1505(1), 012018. https://doi.org/10.1088/1755-1315/1505/1/012018
3. Begum, R. A., Raihan, A., & Said, M. N. M. (2020). Dynamic impacts of economic growth and forested
area on carbon dioxide emissions in Malaysia. Sustainability, 12(22), 9375. https://www.mdpi.com/2071-
1050/12/22/9375
4. Chuah, S. C., Cheam, C. L., Chua, M. S., & […] (2025). The nexus between economic growth, foreign
direct investment, renewable energy and ecological footprint in Malaysia. Institutions and Economies,
17, no. 2 (2025): 57-85. http://jice.um.edu.my/index.php/ijie/article/view/60004
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 4957 www.rsisinternational.org
5. Jamaludin, N. F., Muis, Z. A., Hashim, H., Mohamed, O. Y., & Keng, L. L. (2024). A holistic mitigation
model for net zero emissions in the palm oil industry. Heliyon, 10(6), e27265.
https://doi.org/10.1016/j.heliyon.2024.e27265
6. Lau, E., Tan, C. C., & Tang, C. F. (2016). Dynamic linkages among hydroelectricity consumption,
economic growth, and carbon dioxide emission in Malaysia. Energy Sources, Part B: Economics,
Planning, and Policy. 11(11), 1042-1049 https://doi.org/10.1080/15567249.2014.922135
7. Lau, L. S., Yii, K. J., Lee, C. Y., Chong, Y. L., & Lee, E. H. (2018). Investigating the determinants of
renewable energy consumption in Malaysia: An ARDL approach. International Journal of Business and
Society. 19(3), 886-903 https://www.ijbs.unimas.my/images/repository/pdf/Vol19-no3-paper20.pdf
8. Mentel, G., Tarczyński, W., Azadi, H., Abdurakmanov, K., Zakirova, E., & Salahodjaev, R. (2022). R&D
human capital, renewable energy and CO₂ emissions: Evidence from 26 countries. Energies, 15(23),
9205. https://www.mdpi.com/1996-1073/15/23/9205
9. Naz, S., Sultan, R., Zaman, K., Aldakhil, A. M., Nassani, A. A., & Abro, M. M. Q. (2019). Moderating
and mediating role of renewable energy consumption, FDI inflows, and economic growth on carbon
dioxide emissions: Evidence from robust least square estimator. Environmental Science and Pollution
Research. 26(3), 2806-2819 https://doi.org/10.1007/s11356-018-3837-6
10. Pesaran, M. H., Shin, Y., & Smith, R. P. (1999). Pooled mean group estimation of dynamic heterogeneous
panels. Journal of the American Statistical Association. 94(446), 621-634
https://doi.org/10.1080/01621459.1999.10474156
11. Pham, P. M. H., Nguyen, T. D., Nguyen, M., & Tran, N. T. (2025). FDI inflows and carbon emissions:
New global evidence. Discover Sustainability. 6(1), 432 https://doi.org/10.1007/s43621-025-01292-9
12. Ponce, P., Río-Rama, M. C. del, Álvarez-García, J., & Oliveira, C. (2021). Forest conservation and
renewable energy consumption: An ARDL approach. Forests, 12(2), 255. https://www.mdpi.com/1999-
4907/12/2/255
13. Saboori, B., Sulaiman, J., & Mohd, S. (2016). Environmental Kuznets curve and energy consumption in
Malaysia: A cointegration approach. Energy Sources, Part B: Economics, Planning, and Policy, 11(9),
861–867. https://doi.org/10.1080/15567249.2012.662264
14. Sarkar, M. S. K., Al-Amin, A. Q., Mustapa, S. I., & Ahsan, M. R. (2019). Energy consumption, CO₂
emission and economic growth: Empirical evidence for Malaysia. International Journal of Environment
and Sustainable Development. 18(3), 318-334 https://doi.org/10.1504/IJESD.2019.101036
15. Segar, Y., Akhir, N. H. M., & Sanusi, N. A. (2024). Forecasting CO₂ emissions in Malaysia through
ARIMA modelling: Implications for environmental policy. International Journal of Design & Nature and
Ecodynamics, 19(3), 849–857. https://doi.org/10.18280/ijdne.190315
16. Shahid, S., Minhans, A., & Puan, O. C. (2014). Assessment of greenhouse gas emission reduction
measures in transportation sector of Malaysia. Jurnal Teknologi, 70(4).
https://doi.org/10.11113/jt.v70.3481
17. Songkin, M., & Jaafar, M. Y. H. (2023, March 15). Electric vehicle readiness in Sabah: Overview of
market forecast and adoption challenges. In 2023 6th International Conference on Energy Conservation
and Efficiency (ICECE) (Lahore, Pakistan). IEEE. https://doi.org/10.1109/icece58062.2023.10092507
18. Suman, R., Dwivedi, G., Singh, G., Gidwani, K., & […] (2025). Analyzing the effect of clean cooking
fuel and technologies on carbon emissions in G20 countries. Discover Sustainability. 6(1), 93
https://doi.org/10.1007/s43621-025-00858-x
19. Vaicondam, Y., Mustafa, A. M. A. A., Roslan, S. N. M., Ming, K. L. Y., & Ramayah, M. (2024). GHG
and carbon emission intensity: Examining their impact on financial performance. International Journal
of Energy Economics and Policy, 15(1), 190–196. https://doi.org/10.32479/ijeep.17474
20. Xuan, V. N. (2025). Determinants of carbon dioxide emissions in technology revolution 6.0: New insights
from Brunei. Energy Strategy Reviews. 57, 101633
https://www.sciencedirect.com/science/article/pii/S2211467X24003420
21. Zhao, J. (2025). Exploring the nonlinear dynamic relationship between foreign direct investment,
renewable energy consumption, economic growth, and carbon emissions: A panel threshold VAR
analysis. SAGE Open. 15(2), 21582440251343957 https://doi.org/10.1177/21582440251343957