Growth, Energy Use and CO₂ Emissions in Somalia: What a 2025 ARDL Study Reveals

A digest of Sheikh Ali & Dhiif (2025), International Journal of Economics and Financial Issues

As Somalia's economy recovers, a key question for climate and energy policy is which parts of that growth drive carbon emissions. Research on growth, energy and emissions has focused mainly on larger, more industrialised economies, leaving countries with small industrial sectors and heavy reliance on biomass and fossil fuels less studied. This 2025 study uses national data for 1990–2019 and an autoregressive distributed lag (ARDL) model to estimate how GDP per capita, energy consumption and industrial value added relate to Somalia's CO₂ emissions in the short and long run. It is relevant to policymakers, energy planners and researchers working on low-carbon development in fragile economies.

The study covers 1990–2019, giving 30 annual observations. The long-run GDP elasticity in the ARDL model is 0.81, the long-run energy elasticity is 0.96, and the error-correction term is −0.266 per year.

Article at a Glance

The article is "Impact of Economic Growth, Energy Consumption, and Industrialization on CO₂ Emissions: Evidence from Somalia," by Ali Yassin Sheikh Ali (SIMAD University) and Abdirahman Abdi Dhiif (Skillful Academy). It was published in the International Journal of Economics and Financial Issues, Vol. 15, No. 4, pp. 135–143 (EconJournals; open access, CC BY 4.0), in 2025 (received 4 December 2024; accepted 10 May 2025). Its research area is energy and environmental economics, and its study context is Somalia, using national annual data for 1990–2019 (World Bank, SESRIC). The method is quantitative time series: ADF and PP unit-root tests, an ARDL bounds test and error-correction model, an FMOLS robustness check, and Granger causality. The DOI is https://doi.org/10.32479/ijefi.18928.

The Research Problem

The Environmental Kuznets Curve and related theories suggest emissions first rise with growth, energy use and industrialisation. The authors argue that these links are well studied in larger economies but not in Somalia, where industry is nascent, energy relies heavily on biomass, charcoal and fossil fuels, and environmental regulation is weak — so the usual patterns may not hold.

Research Objective

The study estimates the combined and separate effects of three factors on CO₂ emissions in Somalia, in the short and long run. The first is economic growth, measured by GDP per capita (constant 2015 prices). The second is energy consumption, in kilograms of oil equivalent per person. The third is industrialisation, measured by industry value added.

How the Study Was Conducted

A single-country time-series design with 30 annual observations. CO₂ emissions (kilotons) are the dependent variable; all variables are in natural logarithms, so coefficients read as elasticities.

The analysis followed six steps. In step 01, Unit roots, ADF and PP tests showed energy consumption stationary in levels and the other variables stationary after differencing. In step 02, Bounds test, F = 9.37, above the 1% upper bound (5.61), indicating a long-run relationship. In step 03, ARDL estimation, long-run coefficients and a short-run error-correction model were fitted. In step 04, Diagnostics, normality, serial-correlation and heteroskedasticity tests passed, and CUSUM was stable. In step 05, Robustness, the long-run relationship was re-estimated with FMOLS. In step 06, Causality, pairwise Granger causality tests and an omitted-variables test were run.

Key Findings

The elasticities come from Tables 5 and 7 of the article, and n.s. means not statistically significant. Energy consumption has an ARDL long-run elasticity of 0.956 (p = .026), an FMOLS long-run elasticity of 0.135 (n.s.) and an ARDL short-run elasticity of 0.160 (p = .004). GDP per capita has 0.807 (p = .030) in the ARDL long run, 0.860 (p < .01) in the FMOLS long run and −0.232 (n.s.) in the ARDL short run. Industrialisation has 0.215 (n.s.) in the ARDL long run and −0.063 (n.s.) in the FMOLS long run, with no short-run value reported. The error-correction term is −0.266 (p < .001).

1. Economic growth is consistently linked to higher emissions. A 1% rise in GDP per capita is associated with about 0.8–0.9% more CO₂ in the long run in both ARDL and FMOLS; the short-run effect is not significant.

2. Energy use matters in ARDL, but the result is not robust. ARDL links a 1% rise in energy use to 0.96% more emissions in the long run and 0.16% in the short run, but FMOLS finds no significant long-run effect.

3. Industrialisation has no significant long-run effect. The authors attribute this to Somalia's small industrial base; Granger tests nonetheless suggest two-way causality between industry and emissions. Deviations from equilibrium close at about 27% a year.

Reading note: In Table 5, standard errors, t-values and p-values do not reconcile for GDP (0.807 ÷ 0.136 ≈ 5.9, not the reported 1.58) or industrialisation (t = 2.32 with p = .13). The abstract's energy result is not confirmed by FMOLS, and energy use is essentially uncorrelated with emissions (r = −0.004). The methods section also contains an equation with variables not used in the study (FDI, urban population, renewable energy) and a header from another article.

What the Study Contributes

The contribution is primarily empirical: an updated single-country estimate of how growth, energy use and industry relate to emissions in Somalia. The authors recommend investment in solar, wind and hydro power, energy-efficiency incentives, carbon pricing and stricter fuel standards, low-carbon transport, sustainable industrialisation rules, green research and innovation, and regional energy cooperation.

Important Limitations

Identified by the authors: The paper does not include a dedicated limitations section.

Skilful cautions arising from the study design: 30 annual observations is a small sample, and differenced CO₂ is stationary only at the 10% level. The logged GDP series averages 2.50, which would imply about US$12 per person in natural logs — suggesting a different log base or unit. The Environmental Kuznets Curve is discussed but not tested (no squared GDP term). Finally, data sources are described only briefly, and Somali statistics for the 1990s are of uncertain quality.

Why This Research Matters

Somalia's emissions are small globally, but choices made now about power generation, transport and industry will shape its future emissions path. Evidence on which drivers matter helps target energy and climate policy.

Skilful Research Insight

Editorial commentary — not a finding of the study. Three lessons stand out. First, when a robustness estimator disagrees with the main model, say so prominently: here FMOLS supports the growth result but not the energy result. Second, check that coefficients, standard errors and p-values agree before publication. Third, comparison across studies is revealing — another ARDL study of Somalia for the same years, Nur et al. (2024) on capital formation, found GDP insignificant in its main model. Replicating both with shared data, and testing an EKC directly, would clarify how growth affects Somali emissions.

Read the Original Research

The original article is "Impact of Economic Growth, Energy Consumption, and Industrialization on CO₂ Emissions: Evidence from Somalia," published in the International Journal of Economics and Financial Issues, 15(4), 135–143 (2025), by Ali Yassin Sheikh Ali and Abdirahman Abdi Dhiif. Its DOI is https://doi.org/10.32479/ijefi.18928, 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 "Economic Growth, Energy Use and CO₂ Emissions in Somalia: 2025 Study." The meta description is "A Skilful digest of a 2025 ARDL study of how GDP per capita, energy consumption and industrialisation relate to Somalia's CO₂ emissions, 1990–2019." The URL slug is /economic-growth-energy-co2-emissions-somalia. The primary keyword is CO2 emissions in Somalia. The secondary keywords are economic growth and emissions; energy consumption; industrialisation; ARDL bounds test; FMOLS; environmental Kuznets curve.