Renewable Energy, Domestic Investment, and Environmental Quality: What a 2025 Study Reveals

A digest of Dhiif & Ali (2025), International Journal of Energy Economics and Policy

Somalia's heavy reliance on inefficiently used biomass, alongside largely untapped solar and wind potential, makes it a distinctive test case for whether renewable energy and domestic investment can improve environmental outcomes in a fragile, data-scarce economy. This study estimates the long- and short-run relationships between environmental quality — measured directly as CO2 emissions — and renewable energy consumption, domestic investment, economic growth, population growth and agricultural output in Somalia from 1990 to 2020, using an ARDL bounds-testing approach with FMOLS robustness checks and Granger causality tests. Its central results support a clean-energy story, but its own labelling of "environmental quality" is worth reading carefully before taking the headline claims at face value.

The study covers 1990–2020, giving 31 annual observations in a single-country ARDL design, and uses 2 estimators: the ARDL bounds test and FMOLS as a robustness check. In the ARDL model, each 1% rise in renewable energy is linked to a 14.18% change in CO2, and the R² of the ARDL model is 94.7%.

Article at a Glance

The article is "Assessing the Influence of Renewable Energy Consumption and Domestic Investment on Environmental Quality in Somalia," by Abdirahman Abdi Dhiif (Skilful Academy, Somalia) and Ali Yassin Sheikh Ali (Faculty of Economics, SIMAD University, Somalia). It was published in the International Journal of Energy Economics and Policy, Vol. 15, No. 4, pp. 576–587 (EconJournals; open access, CC BY 4.0), in 2025 (received 1 February 2025; accepted 26 May 2025). Its research area is energy and environmental economics, and its study context is Somalia, using national annual data for 1990–2020 (31 observations). The method is time series: ARDL bounds testing with an error-correction model, FMOLS robustness, and pairwise Granger causality. The DOI is https://doi.org/10.32479/ijeep.19603.

The Research Problem

Environmental degradation in Somalia is compounded by weak infrastructure, economic instability and political insecurity, with continued dependence on conventional biomass driving deforestation, air pollution and carbon emissions. At the same time, Somalia's largely unexploited solar and wind resources could, if harnessed, ease environmental strain while supporting economic development — but inadequate regulation, limited investment and thin infrastructure have held back that transition. The authors argue that while economic growth's link to environmental quality is well studied, research on domestic investment's role — alone and jointly with renewable energy — remains scarce for Somalia specifically.

Research Objective

The study asks three linked questions: what impact does renewable energy consumption have on environmental quality in Somalia; what is the significance of domestic investment for environmental sustainability; and how do the two combine to shape long-term environmental outcomes? Environmental quality (EQ) is proxied directly by CO2 emissions (in kilotons); renewable energy consumption (REC), domestic investment (DI, via gross fixed capital formation), economic growth (GDP per capita) and population growth (POG) are modelled as its determinants, with agricultural output (AGP) as an additional control. All variables are logged.

How the Study Was Conducted

Annual Somalia data for 1990–2020 (31 observations) were drawn from the World Development Indicators and the Organization of Islamic Cooperation. With a mix of stationary and non-stationary series, the authors use the ARDL bounds-testing approach, which accommodates both I(0) and I(1) variables without requiring a common integration order, to estimate long-run elasticities and short-run error-correction dynamics.

The analysis followed six steps. In step 01, Stationarity, ADF and PP tests showed that GDP, population growth and renewable energy are stationary at levels, while CO2, domestic investment and agricultural output require first differencing. In step 02, Cointegration, the ARDL bounds test gave a Wald F-statistic of 24.15, well above the 1% upper bound of 4.68, confirming a long-run relationship. In step 03, Estimation, ARDL long-run coefficients were estimated along with an error-correction model (ECM) for short-run dynamics, with an ECT of −0.678. In step 04, Diagnostics, the Jarque-Bera, Breusch-Godfrey and Breusch-Pagan tests were all non-significant, supporting normality, no serial correlation and homoscedasticity. In step 05, Robustness, Fully Modified OLS (FMOLS) was used to re-estimate the long-run relationship as a cross-check on the ARDL results. In step 06, Causality, pairwise Granger causality tests were run across CO2, GDP, population growth, renewable energy, domestic investment and agricultural output.

Key Findings

The long-run coefficients on CO2 are reported for the ARDL and FMOLS estimators, as shown in Tables 6 and 8 of the article. Significance is marked at 5% () and 1% (), and negative values mean the variable is associated with lower CO2 emissions. Domestic investment has −0.099** in ARDL and −0.071** in FMOLS. Economic growth (GDP) has −0.288*** in ARDL and 0.084 in FMOLS (not significant). Renewable energy has −14.181*** in ARDL and −6.582*** in FMOLS. Agricultural output has −0.002 in ARDL and −0.034 in FMOLS, neither significant. Population growth has 1.898*** in ARDL and 0.535** in FMOLS.

1. Renewable energy shows by far the largest emissions-reducing effect — but the two estimators disagree sharply on size. A 1% rise in renewable energy consumption is linked to a 14.18% fall in CO2 emissions in the ARDL model, and a 6.58% fall in the FMOLS robustness check. Both are highly significant and both support the same story, but the point estimates differ by roughly a factor of two — a gap the paper does not reconcile.

Reading note: The paper defines "Environmental Quality" (EQ) as CO2 emissions themselves, not an inverted index. So when the text describes population growth as having "a sustained positive link with environmental quality," it means population growth raises CO2 emissions — the opposite of what "positive for environmental quality" would normally suggest. Every directional claim about EQ in this paper should be read as a claim about emissions levels, not conventional environmental quality.

1. Domestic investment consistently reduces CO2 emissions, in both the long run and short run. Domestic investment carries a negative, significant coefficient in both ARDL (−0.099) and FMOLS (−0.071), and also has a beneficial short-run effect (−0.043) in the error-correction model. The authors read this as evidence that Somalia's domestic investment, plausibly channelled toward infrastructure and cleaner practices, is emissions-reducing rather than emissions-driving.

2. Economic growth's effect on emissions is not robust across models. GDP raises long-run CO2 emissions in ARDL (+0.288%, consistent with an early-stage Environmental Kuznets Curve reading) but is small and statistically insignificant in FMOLS (0.084). The paper treats the ARDL result as the headline finding without flagging that its main robustness check does not confirm it.

3. Population growth raises emissions, and the paper reads this as a positive, if paradoxical, environmental result. Population growth has a strongly positive coefficient on CO2 in both the long run (ARDL +1.898, FMOLS +0.535) and the short run (+4.056). Because the paper's EQ variable is CO2 itself, rising emissions with population growth is repeatedly described as an unexpected environmental "benefit" — language that only makes sense once EQ is understood as raw emissions rather than emissions-adjusted quality.

4. Granger causality points to a two-way, interlinked system rather than one dominant driver. Renewable energy consumption Granger-causes GDP (not the reverse), GDP Granger-causes both CO2 emissions and domestic investment, and domestic investment and population growth Granger-cause each other bidirectionally. The authors read this web of causal links as evidence that economic, demographic and environmental factors in Somalia move together rather than in a single direction.

What the Study Contributes

The contribution is empirical: a Somalia-specific ARDL analysis of renewable energy and domestic investment together, a combination the authors argue has been studied separately elsewhere but not jointly for Somalia. Diagnostic and robustness checks (FMOLS, Jarque-Bera, Breusch-Godfrey, Breusch-Pagan) and pairwise Granger causality tests add texture beyond a single long-run regression, and the paper translates its estimates into concrete policy recommendations — subsidies, tax incentives and regulatory frameworks to expand renewable energy and channel domestic investment toward green infrastructure.

Important Limitations

Identified by the authors: No explicit limitations section is given in the article; the conclusion instead points to future work on government, education and foreign-assistance channels not covered by the current model.

Skilful cautions arising from the study design: 31 annual observations is a thin sample for an ARDL/ECM model with five explanatory variables plus multiple lags — typical of single-country time series in data-scarce settings, but a real constraint on statistical power. "Environmental Quality" is operationalised as raw CO2 emissions with no inversion, so the paper's own "positive/negative effect on EQ" language runs opposite to an intuitive reading throughout the results and discussion (see Reading Note). The renewable-energy elasticity is roughly twice as large in ARDL (−14.18%) as in FMOLS (−6.58%); both are treated as confirming the same result, but the two-fold gap matters for anyone using the estimate for policy calculations. Pairwise Granger causality tests run on only 29 observations per pair, below the sample sizes typically recommended for reliable causal inference. Finally, GDP is significant in ARDL but not in FMOLS, a genuine model-sensitivity that the paper reads only as general robustness rather than flagging the discrepancy.

Why This Research Matters

Somalia is actively expanding solar and wind capacity as part of its development strategy, and this study's elasticities — however wide their range across estimators — give planners and donors, including the World Bank and regional partners cited in the paper, a concrete starting point for weighing renewable energy and domestic investment as emissions-reduction levers. For researchers, the ARDL-plus-FMOLS-plus-Granger-causality design is a compact, replicable template for similar questions in other fragile, data-scarce economies.

Skilful Research Insight

Editorial commentary — not a finding of the study. Three things stand out. First, always check how a paper defines its outcome variable before reading its directional language: here "environmental quality" is raw CO2 emissions, not an inverted quality index, so the population-growth finding reads as "positive for environmental quality" only under that framing — a general reader would otherwise assume the opposite. Second, cross-check magnitude as well as sign across estimators: the two-fold gap between ARDL (−14.18%) and FMOLS (−6.58%) on renewable energy matters for any downstream calculation, even though both point the same direction. Third, with only 31 observations feeding a multi-variable ARDL model with short-run lags, these elasticities are best read as suggestive orders of magnitude for Somalia's context rather than precise, exportable multipliers.

Read the Original Research

The original article is "Assessing the Influence of Renewable Energy Consumption and Domestic Investment on Environmental Quality in Somalia," published in the International Journal of Energy Economics and Policy, 15(4), 576–587 (2025), by Abdirahman Abdi Dhiif and Ali Yassin Sheikh Ali. Its DOI is https://doi.org/10.32479/ijeep.19603, and the publisher/journal page is EconJournals, www.econjournals.com (open access, CC BY 4.0). Skilful summarises; all findings belong to the original authors.

SEO Information

The SEO title is "Renewable Energy & Domestic Investment in Somalia: 2025 Study Summary." The meta description is "A Skilful digest of a 2025 IJEEP ARDL study on how renewable energy consumption and domestic investment relate to CO2 emissions in Somalia." The URL slug is /renewable-energy-domestic-investment-environmental-quality-somalia. The primary keyword is renewable energy domestic investment environmental quality Somalia. The secondary keywords are ARDL bounds test; Granger causality; CO2 emissions Somalia; FMOLS robustness; Environmental Kuznets Curve.