Factors influencing Somalia household’s willingness to pay renewable energy: employing structural equation modeling

Abstract

This study used PLS path analysis and structural equation modelling (SEM) to examine the factors influencing Somali households’ willingness to pay for renewable energy. Non-probability purposive sampling was used in a quantitative survey to select respondents from Mogadishu, the capital city of Somalia. 300 home power bill payers who were informed about energy costs and renewable energy requirements were given a standardized closed-ended questionnaire. Following data cleaning, SPSS version-25 and SmartPLS-4 were used to analyze 255 valid replies using descriptive and inferential statistics. With an R2 of 0.428, the structural model has moderate explanatory power, accounting for 42.8% of the variance in willingness to pay for renewable energy. The model’s robustness is confirmed by an adjusted R2 of 0.402. The findings showed that consumer intention, environmental concern, perceived behavioral control, subjective norms, and Attitude have a positive significant impact on willingness to pay for renewable energy. Belief about the cost of renewable energy shows no significant relationship with willingness to pay for renewable energy. The results of the moderation analysis indicate that the relationships between environmental concern, subjective norms, and attitude with willingness to pay for renewable energy are considerably moderated by customer intention. However, the relationship between perceived behavioral control and belief about the cost of renewable energy with willingness to pay for renewable energy is not moderated by consumer intention. The findings offer policymakers and renewable energy stakeholders insights to increase adoption rates.

Introduction

Climate change and environmental problems has received increasing attention in political and economic discussions during the past decades¹. Using traditional solid fuel for cooking and home heating contributes to indoor air pollution, responsible for about 3.2 million deaths in 2020². Therefore, using clean energy for household consumption may reduce indoor air pollution and household members’ health complications. One of the most important initiatives to combat climate change is to reduce energy consumption^3,4. During the last few decades, electricity consumption has grown rapidly, mainly in the residential and service sectors⁵. This rapid electricity consumption has led to increasing CO2 emissions and ultimately impacts global warming^4,6. The demand for energy is at its peak, driven by population growth and economic development⁷. Therefore, policymakers are considering alternative energy sources and ways to reduce carbon footprints by relying less on fossil fuels¹. In turn, lower fossil fuel use and fewer greenhouse gas emissions can be achieved⁸. The installation of energy-efficient appliances (EEAs) plays a significant role in reducing household energy consumption⁴. However, the lived experience of climate shocks can accelerate adoption of renewable energy⁹.

The accessibility of electricity can raise a family’s standard of living². Energy is crucial in all human endeavors, such that the growth and development of any economy are hinged on it¹⁰. Hence, the people need energy for various purposes including lighting, cooking, transportation and even entertainment. The rising demand for energy and the risk of fossil fuel depletion have driven the rapid development of renewable energy sources to meet clients’ needs⁵. Renewable energy refers to energy derived from natural sources such as wind, sunlight, water, and biomass^11,12. Therefore, environmentally-friendly products are demanded by consumers as a new segment to protect against climate change¹³. Moreover, consumers who engage in pro-environmental behavior (PEB) have a lower negative environmental impact^14,15. Pro-environmental consumer behavior research is conducted in developed markets but is still in its early stages in several emerging markets, including Asia and Africa¹⁶. Most households choose clean energy, for which they are ready to spend a portion of their monthly budget¹⁷. A considerable majority of the population in most rural households in developing countries, such as Sub-Saharan Africa, lacks electricity. In 2020, only about 48.4% of the African population had access to electricity².

However, Africa still struggles with a persistent energy crisis despite its abundant energy sources and potential for RE growth, with more than $10 billion being spent each year on kerosene lighting in Sub-Saharan Africa¹⁸. In 2019, the overall energy produced in all varieties amounted to 245 GW¹⁹. However, interest in renewable energy (RE) options is rising due to their ability to serve as a cleaner alternative to fossil fuels, which come with high production costs, contribute to environmental damage, and pose health risks to humans and animals²⁰. This study conducted in Kenya indicated that while renewable uptake remains low, but physical accessibility to energy infrastructure significantly influences a household’s willingness to diversify their fuel choices²¹. willingness to pay (WTP) is defined as the ability of consumers to pay for a particular product or service, which is a vital tool to measure that commodity’s value or demand in the market²².

There are few studies conducted in Somalia about energy transition from fossil fuel to clean energy. Therefore, Somalia remains understudied for understanding household energy transitions from fossil fuel to clean energy, particularly, primary factors that influence WTP renewable energy based on the TPB theory. A previous study examined Somali SMEs and found that investment intention is driven not only by Theory of Planned Behavior (TPB) constructs but also by risk aversion, suggesting that perceived risk is a vital component of energy decision-making in the region²³. However, that study ignores important constructs in Somali context such as beliefs about the costs of renewable energy and environmental concern. Hence, this study includes those constructs. Moreover, this study also fills the gap by testing consumer intention as a moderator variable between TPB constructs (attitude, subjective norms, perceived behavioral control) with additional two more constructs beliefs about the costs of renewable energy and environmental concern on WTP, which would be a methodological innovation.

The following are the specific objectives:

• To study the factors influencing willingness to pay renewable energy in Somalia.

• To extend the Theory of Planned Behavior by incorporating two more variables: Beliefs about the costs of renewable energy and environmental concern.

• To investigate the moderating role of consumer intention.

Renewable energy in Somalia

The momentum transition from fossil fuel energy to renewable energy in Somalia is growing, driven by the need to shift from expensive, inefficient diesel generators to more sustainable alternatives. The nation lacks a unified national grid, and a significant portion of its energy consumption depends on firewood and charcoal, which exacerbate environmental degradation. Given its prevalence in Somalia, solar energy is emerging as a feasible alternative. A multitude of studies have investigated the potential and economic viability of solar power systems, both off-grid and hybrid, for meeting the nation’s energy requirements. These initiatives constitute an integral part of a comprehensive strategy to enhance the electrification rate while simultaneously reducing dependence on imported diesel. Somalia is characterized by an annual solar insolation of approximately 3,000 h, with daily solar radiation metrics fluctuating between 5 and 7 kWh/m², thereby rendering it exceptionally suitable for the generation of solar energy²⁴.

Hybrid systems combining solar PV, diesel generators, and battery storage have been explored to optimize costs and reduce environmental impacts. The hybrid PV-DG-grid system without battery storage was found to be the most cost-effective²⁵. In Beledweyne, a hybrid system including solar, wind, and hydropower was analyzed, highlighting the potential for diverse renewable energy solutions²⁶.

Rural electrification remains a challenge due to low population density and economic constraints, but solar-powered mini-grids offer a promising solution²⁷. The transition to renewable energy is crucial for reducing Somalia’s dependence on imported diesel and mitigating environmental impacts²⁵. Although the integration of renewable energy sources in Somalia is advancing, persistent obstacles, including financial constraints and the need for infrastructure development, persist. Nevertheless, the country’s abundant solar energy potential and the continuation of related initiatives signal an optimistic pathway towards sustainable energy solutions.

Somalia faces significant energy challenges, with limited grid infrastructure and widespread reliance on expensive, polluting energy sources. Understanding consumer behavior and willingness to adopt renewable energy is crucial for developing effective energy policies and deployment strategies. While existing studies have examined various aspects of renewable energy investment and adoption, there are specific areas that remain underexplored, particularly in the context of Somalia^23,28. A study on the adoption of renewable energy in Somalia reveals several gaps that need to be addressed to enhance understanding and facilitate the transition to sustainable energy sources.

Theory of planned behavior (TPB) and hypothesis development

The Theory of Planned Behavior (TPB) is used to explain behaviors in which individuals can exercise self-control²⁹. The theory uses various constructs to predict individuals’ control over their behavior, including attitude, subjective norms, and perceived behavioral control. The TPB is a widely used model to predict consumer intentions, including the adoption of renewable energy. In the context of Somalia, the model has been extended to incorporate two additional variables: beliefs about the costs of renewable energy and environmental concern. These additions aim to provide a more comprehensive understanding of consumer intentions in adopting renewable energy. No study in Africa has combined beliefs about renewable energy costs and environmental concern within a Theory of Planned Behavior (TPB) model tailored to the African context. Research in Pakistan includes both factors, but it is not set in Africa. Nigerian studies address costs but do not explicitly add environmental concerns to their models. Moreover, previous studies have not modeled consumer intention as a moderating variable in Theory of Planned Behavior (TPB) research on renewable energy; instead, intention is typically treated as an outcome or a proximal predictor^5,23,30,31. However, this study uses consumer intentions as a moderating variable between attitude, subjective norms, perceived behavioral control, belief about the costs of renewable energy, environmental concern, and WTP for renewable energy.

The attitude in TPB is defined as an individual’s evaluation of a particular behavior as favorable or unfavorable³². Consumers with favorable views on renewable energy’s benefits are more likely to support and adopt it³³. This study reports a positive correlation between attitude and consumer intention to purchase environmentally friendly vehicles³⁴. Similarly, this study exposed that attitude is positively related to consumer intention to utilize less energy. Attitude is a dominant predictor of residential energy consumption⁴. Attitudes are shaped by perceived utility and social influences, which can enhance consumer commitment to renewable energy adoption³⁵. Positive attitudes, shaped by beliefs about the benefits of renewable energy, lead to a higher likelihood of consumers being willing to adopt for green energy options³⁶. Although, attitudes towards renewable energy investments are identified as a determinant of investment intentions, there is limited study on how these attitudes specifically influence consumer willingness to adopt renewable energy in Somalia²³. The relationship between attitude and behavioral intention is strong, but further exploration is needed to understand the nuances of this relationship in the Somali context³⁷.

H1: Attitude positively influences consumer WTP for renewable energy in Somalia.

H10: Consumer intention to use renewable energy positively moderates between attitude and WTP for renewable energy.

Subjective Norms are perceived social stress of performing or not performing a particular behavior is termed the subjective norm (SBN)³⁸. Subjective norms —the perceived social pressure to engage in renewable energy consumption — positively influence attitudes and WTP. Consumers are more likely to invest in renewable energy if they believe their peers support such actions^36,39. Generally, the actions and opinions of other people have a great impact on consumers’ buying intentions³³. The subjective Norms are recognized as influencing investment intentions, yet there is a lack of detailed analysis on how societal pressures and cultural expectations impact individual willingness to adopt renewable energy²³.

H6: Subjective Norms positively influence consumer WTP for renewable energy in Somalia.

H9: Consumer intention to use renewable energy positively moderates between subjective Norms and WTP for renewable energy.

Perceived Behavioral Control (PBC) refers to an individual’s belief in their ability to perform a specific behavior, which directly influences their intentions and indirectly affects their actual behavior³². These studies found that PBC positively impacts consumer intention to conserve energy^40,41. The PBC plays a critical role in influencing consumers’ decisions to purchase energy-efficient vehicles, highlighting its importance in broader sustainable technology adoption⁴². Although perceived behavioral control is acknowledged as a factor in renewable energy investment, its specific impact on consumer adoption intentions in Somalia remains underexplored³⁷.

H5: perceived behavioral control positively influences consumer WTP for renewable energy in Somalia.

H11: Consumer intention to use renewable energy positively moderates between perceived behavioral control and WTP for renewable energy.

Belief about Renewable Energy Cost (BREC) refers to consumer perception of the financial sacrifices associated with adopting renewable power technologies (RPTs). This variable plays a critical role in shaping consumer behavior, as the high upfront costs and maintenance expenses of renewable energy systems often act as barriers to adoption. Perceptions of renewable energy costs significantly influence consumer adoption intentions. High perceived costs can deter adoption, while perceived affordability can encourage it⁴³.

When consumers hold strong beliefs about the advantages of renewable energy, they are more inclined to develop favorable attitudes towards it, which in turn enhances their WTP³⁶. However, beliefs about the costs of renewable energy can negatively impact the adoption of renewable energy. If consumers perceive renewable energy as expensive, their WTP decreases³³. Belief About Costs The economic analysis of solar systems in Somalia highlights cost-effectiveness, but there is insufficient research on how consumer beliefs about these costs affect their adoption decisions⁴⁴. The financial implications of renewable energy adoption, such as initial investment and payback periods, need further investigation to understand consumer perceptions⁴⁴.

H2: Belief about the cost of renewable energy positively influences consumer WTP for renewable energy in Somalia.

H7: Consumer intention to use renewable energy positively moderates between Belief about Renewable Energy Cost and WTP for renewable energy.

Environmental Concern (ECN) refers to the level of awareness individuals have of environmental problems and their motivation to address them. It is often considered a key factor influencing consumer adoption of renewable energy (RE) technologies. While environmental awareness is generally associated with positive attitudes towards renewables, it does not consistently predict the adoption of renewable energy⁴⁵. Environmental awareness and concern are crucial in shaping attitudes towards renewable energy, influencing both the intention and actual adoption of green technologies⁴⁶. The lack of government-led awareness campaigns and education on ecological issues may further diminish ECN’s role in purchase decisions. Similarly, the environmental benefits of renewable energy are noted, but there is a gap in understanding how environmental concerns drive consumer adoption in Somalia²⁷. However, the WTP for renewable energy varies by type, indicating that consumer preferences are not uniform across different renewable sources⁴⁷.

H4: Environmental Concern positively influences consumer WTP for renewable energy in Somalia.

H8: Consumer intention to use renewable energy positively moderates the relationship between Environmental Concern and WTP for renewable energy.

H3: Consumer intention to use renewable energy positively influences consumer WTP for renewable energy in Somalia (Figs. 1, 2).

Fig. 1.

Conceptual framework.

Fig. 2.

Flowchart model.

Research methods

Method overview

This study uses a systematic approach to examine the factors that influence household WTP for renewable energy in Somalia. The method is intended to assure the acquisition of trustworthy and valid data, while adhering to research ethics. It incorporates suitable research design, sample processes, data gathering methods, and analytical methodologies to achieve the study objectives. The technique also allows for the investigation of relationships between key variables as well as the offering of empirically supported conclusions to back up the findings of the study.

Research instrument

The method utilized for this study is a survey and regression research design based on quantitative methodologies. The researchers chose the sample using a nonprobability sampling approach, specifically purposive sampling. A structured, closed-ended questionnaire was used to collect data from residents of Mogadishu, Somalia’s capital city. Mogadishu was chosen since it is the most populous city in Somalia, and people pay energy bills. The study focused on persons who pay household power bills who are knowledgeable about the cost of electricity and the need for renewable energy. This study’s sample size was 300 respondents; however, after data cleaning, only 255 were considered credible for analysis. Researchers used structured questionnaires. Data were collected using a structured questionnaire administered online (WhatsApp, email) and in person. Data generated for the study were analyzed using descriptive and inferential statistics via the Statistical Package for the Social Sciences (SPSS) version 25 and SMART PLS-4. The study uses descriptive statistics to summarize the collected data and regression analysis to show the impact of the independent variable on the dependent variable.

Measurement scale and SEM model

To develop the research constructs used in the study, various previous works were consulted. All scale items are adopted from⁴⁸. All item scales were measured using the 5-point Likert scale (1 = strongly agree to 5 = strongly disagree) in the descriptive section. Structural equation modeling (SEM) is a widely adopted technique to scrutinize the relationship between different variables⁴⁸. SEM was considered a suitable model for the study because it provides accurate and meaningful outcomes regarding the study constructs⁴⁹. The technique takes measurement error into account by adding measurement error variables that relate to the measurement error segments of observed variables. Therefore, the results of the association between.

constructs are unbiased and reliable⁵⁰. SEM is also used to evaluate the model’s measurement and structural components simultaneously, with bootstrapping techniques used to determine the importance of both direct and indirect effects⁵¹. Besides, the technique is capable to test complex relationships and a multitude of hypotheses simultaneously including group comparisons and mean structures which may be not possible in other techniques and models⁵².

Results and analysis

Descriptive section

Table 1 shows the demographics of this research. 48.6% of the participants were men, while 51.4% were women. The bulk of respondents (46.3%) were between the ages of 21 and 25, while the next-largest group was 18- to 20-year-olds, accounting for 32.2%. Regarding education, 54.5% of respondents were undergraduate students. The bulk of respondents (76.5%) reported being single. The majority of respondents’ households had seven or more individuals, accounting for 56.6% of the total, indicating that Somalis had high household sizes. In terms of income source, the majority of families earned 51.0% from business, with employment accounting for 35.7%. Finally, the monthly energy cost for the majority of Somali families (47.5%) is between $11 and $20.

Table 1.

Respondents profile.

Variable	Frequency	Percent (%)
Gender
Male	124	48.6
Female	131	51.4
Total	255	100.0
Age
18–20 years	82	32.2
21–25 years	118	46.3
26–30 years	15	5.9
31–35 years	8	3.1
Above 35 years	32	12.5
Total	255	100.0
Level of education
Graduate	83	32.5
Undergraduate	139	54.5
High school	17	6.7
Other	16	6.3
Total	255	100.0
Marital status
Married	60	23.5
Single	195	76.5
Total	255	100.0
Household size
1–3 members	43	16.9
4–6 members	65	25.5
7 or more members	147	57.6
Total	255	100.0
Primary source of income
Agriculture	12	4.7
Livestock	22	8.6
Business	130	51.0
Employment	91	35.7
Total	255	100.0
Cost of electricity per month
1–10 $	81	31.8
11–20 $	121	47.5
21–30 $	37	14.5
31–40 $	12	4.7
Above 40 $	4	1.6
Total	255	100.0

Measurement model

This study adopts a quantitative approach and uses structural equation modelling. The measurement model was assessed using factor loadings, composite reliability, average variance extraction, and the Variance Inflation Factor. According to⁵³, to achieve convergent validity, published recommendations require factor loadings of 0.70 or higher, composite reliability of 0.70 or higher, and an average variance extracted of 0.50 or higher. Using the Smart PLS-4 technique, researchers found that all factor loadings were above the suggested 0.70, except for S4, B5, E2, W4, and C6, as indicated in Table 2. However, factor loadings between 0.40 and 0.70 are regarded as acceptable if the average variance extracted is more than 0.50, as stated by (Byrne, 2016). The average variance extracted in this study exceeds 0.50. Furthermore, the composite reliability (CR) values in this study are above the required level of 0.70, as stated by prior studies like^54,55. Regarding the Variance Inflation Factor (VIF), all indicators have values below 5, indicating no multicollinearity concerns. Overall, the model exhibits high explanatory power and predictive relevance, with P2 standing out as the most relevant predictor (see in Table 2).

Table 2.

Loading and reliability factors.

Latent variables	Indicators	Loading	CR	AVE	VIF
Attitude			0.791	0.654
It is wise to utilize renewable power generation technologies (RPTs) at my home	A3	0.817			1.106
It would be pleasant to utilize renewable power generation technologies (RPTs) at my home	A4	0.801			1.106
Subjective norms			0.767	0.524
People who are valuable for me have the opinion that I should use renewable power generation technologies (RPTs) at my home	S1	0.736			1.095
I will use renewable power generation technologies (RPTs) if my friends also use it	S4	0.683			1.202
I will use renewable power generation technologies (RPTs) if my relatives advise me to use	S5	0.750			1.209
Perceived behavioral control			0.807	0.583
I have the required knowledge to utilize renewable power generation technologies (RPTs)	P2	0.795			1.326
I have the resources to utilize renewable power generation technologies (RPTs)	P3	0.772			1.341
I have full control of utilizing renewable power generation technologies (RPTs)	P4	0.721			1.173
Belief about costs of renewable energy utilization			0.773	0.532
Electricity price will be increased, as renewable energy (RE) projects require a high initial capital	B2	0.708			1.144
The utilization of renewable energy (RE) requires high installation costs	B4	0.787			1.223
Battery replacement in case of using solar PV requires additional costs	B5	0.689			1.143
Environmental concern			0.678	0.525
I am worried about climate change	E2	0.545			1.003
Renewable power generation technologies (RPTs) would improve the environment	E4	0.868			1.003
Willingness to Pay (WTP) for Renewable Energy			0.787	0.554
Energy-saving behavior motivate me to pay for renewable energy (RE)	W2	0.704			1.212
Environmental friendliness nature of renewable energy (RE) motivate me to pay for renewable energy (RE)	W3	0.829			1.317
Reliability of renewable power generation technologies (RPTs) motivate me to pay for renewable energy (RE)	W4	0.691			1.152
Consumer’s Intention to Use Renewable Energy			0.792	0.561
I intend to use renewable power generation technologies (RPTs)  if banks provide loan and financial assistance	C4	0.810			1.286
I am planning to spend more money on renewable power generation technologies (RPTs)  than traditional electricity	C5	0.746			1.216
I have a positive intention that other people should utilize renewable power generation technologies (RPTs)  at their homes	C6	0.685			1.168

Discriminant validity was assessed using the Fornell-Larcker criterion. Discriminant validity is achieved when the square root of each construct’s average extracted variance is greater than the correlations between that construct and other constructs⁵⁴. Table 3 shows that the diagonal components correspond to the square roots of each construct’s average retrieved variance. These findings were much higher than the previously reported correlations, demonstrating that the assessments exhibit discriminant validity.

Table 3.

Fornell-Larcher criterion analysis for checking discriminant validity.

A	B	C	E	P	S	W
A	0.809
B	0.253	0.729
C	0.340	0.293	0.749
E	0.250	0.285	0.472	0.725
P	0.249	0.185	0.256	0.136	0.763
S	0.397	0.267	0.322	0.289	0.269	0.724
W	0.340	0.296	0.445	0.458	0.376	0.421	0.744

Discriminant validity testing has become a commonly accepted prerequisite for studying links between reflectively rated constructs. The HTMT ratio criteria were used to assess discriminant validity. Such assessments analyse the distinctness of different concepts within the measuring model⁵⁶. The proposed measuring the HTMT correlation ratio to a threshold of less than 0.85 to determine discriminant validity⁵⁷. However, as demonstrated in Table 4, several HTMT values above the conservative criterion of 0.85. Prior studies imply that a threshold of 0.90 is acceptable, particularly in social science studies including related constructs^57,58. Therefore, most constructs exhibit appropriate discriminant validity. However, a few HTMT values more than 1.0 show possible overlap between constructs, implying that discriminant validity may not be fully proven in these circumstances. This limitation is acknowledged, and future research is recommended to refine the measurement model.

Table 4.

Heterotrait-monotrait (HTMT) ratio.

	A	B	C	E	P	S	W
A
B	0.648
C	1.115	0.504
E	0.450	1.242	0.743
P	0.769	0.314	0.417	0.570
S	0.641	0.475	0.576	1.220	0.451
W	0.493	0.511	0.727	1.735	0.609	0.717

Structural model

Table 5 presents the structural model assessment using key indicators, including R² and Adjusted R². The R² value for WTP for renewable energy is 0.428, suggesting that 42.8% of the variation is explained by independent variables. The modified R² of 0.402 shows the model’s robustness given the number of predictors (Table 6). The findings reported in Table 7 investigate the elements that influence consumer decisions to use renewable energy. The findings are based on hypothesis testing, most often using structural equation modelling (SEM) or regression analysis. This study proposed 11 hypotheses to investigate the correlations between the variables under examination. The structural model results illustrated in Fig. 3 and documented in Table 7 provide useful insights into the linkages between the constructs, offering light on the underlying dynamics driving the events being studied. The PLS path analysis examines direct and indirect effects equally and is the only approach suitable for mediation study. In the structural model, the researchers used p-value and t-value to assess the proposed hypotheses. If the t-value exceeds 1.96 or the p-value is less than 0.10, the hypotheses can be accepted. In this study, a 10% significance level (α = 0.10) was used. While a 0.05 threshold is commonly used, a 0.10 significance level is considered acceptable in exploratory PLS-SEM studies to identify potentially meaningful relationships, particularly in prediction-oriented research and theory development that is still evolving⁵⁹.

Table 5.

Goodness of fit.

Constructs	R-square	R-square adjusted
W	0.428	0.402

Table 6.

Model fit indicators.

Particulars	Saturated model	Estimated model
SRMR	0.091	0.091
d_ULS	1.567	1.565
d_G	0.451	0.449
Chi-square	685.114	683.523
NFI	0.307	0.308

Table 7.

Empirical results of structural model analysis (hypothesis testing).

Hypothesized path	Standardized coefficients	T-value	P-value	Decision
H1. A -> W	0.064	1.510	0.066	Accept
H2. B -> W	0.058	1.231	0.109	Reject
H3. C -> W	0.069	3.001	0.001	Accept
H4. E -> W	0.068	4.102	0.000	Accept
H5. P -> W	0.056	3.662	0.000	Accept
H6. S -> W	0.064	2.448	0.007	Accept
H7. B x C -> W	0.048	0.025	0.490	Reject
H8. E x C -> W	0.062	1.937	0.026	Accept
H9. S x C -> W	0.066	1.767	0.039	Accept
H10. A x C -> W	0.052	1.379	0.084	Accept
H11. P x C -> W	0.068	1.041	0.149	Reject

Fig. 3.

Estimated relationships of structural model.

Table 7 summarizes the hypothesized relationships between the constructs and the dependent variable. H1 supports a relationship between attitude and WTP for renewable energy, with β = 0.064, p < 0.066, and a t-statistic of 1.510. The mindset has a favorable impact on Somali households’ WTP for renewable energy. Therefore, customers have positive attitudes towards renewable energy increase the likelihood that they will pay for it. H2 is not supported, indicating no significant relationship between beliefs about renewable energy costs and WTP for renewable energy (β = 0.058, p < 0.109, and statistics at 1.231). The willingness of Somali households to pay for renewable energy is not significantly influenced by their beliefs about its cost. H3 demonstrates a correlation between consumer intention and WTP for renewable energy, with β = 0.069, p < 0.001, and a t-statistic of 3.001. Consumer intention has a favorable impact on Somali households’ WTP for renewable energy. Consequently, customers plan to spend more on renewable energy than on traditional electricity. Furthermore, customers encourage others to use renewable power-generating technologies (RPTs) at home. H4 demonstrates a relationship between environmental concern and WTP for renewable energy, supported by β = 0.068, p < 0.000, and a t-statistic of 4.102. Environmental concerns positively affect Somali households’ WTP for renewable energy. Therefore, if people are concerned about climate change, they will choose renewable energy over conventional power, thereby benefiting the environment. H5 supports the association between perceived behavioral control and WTP for renewable energy, with β = 0.056, p < 0.000, and a t-statistic of 3.662. perceived behavioral control positively affects Somali households’ WTP for renewable energy. Consequently, if customers have the necessary knowledge and resources, they will use renewable energy, which encourages them to pay for it and gives them complete control over the use of renewable power production technologies (RPTs). H6 demonstrates a correlation between subjective Norms and WTP for renewable energy (β = 0.064, p < 0.007, t-statistic = 2.448). Subjective norms favorably impact Somali households’ WTP for renewable energy. Therefore, if those important to customers, such as friends and family, have a positive attitude towards renewable energy, consumers will adopt renewable power production technologies (RPTs) in their homes.

In terms of moderation, consumer intention moderates the relationship between the findings of H8, H9, and H10 (environmental concern, subjective norms, and attitude) and Somali households’ WTP for renewable energy. While consumer desire to utilize renewable energy did not moderate the relationship between the findings of H7 and H11 (belief about renewable energy cost and perceived behavioral control) and Somali households’ WTP for renewable energy (see in Table 7).

Model fit indicators

The standard root mean square (SRMR) is a measure use to assess the goodness of fit (GoF) in PLS models⁶⁰. SRMR refers to the discrepancies that remain between the correlated data of the samples and the expected correlated model, as described by⁶¹. The SRMR scale ranges from 0 to 1.0, with values close to zero suggesting an ideal fit for the model. A well-fitting model should have an SRMR value that is less than or equal to 0.05, as stated by⁶¹. The SRMR values of 0.091 is somewhat higher than the 0.08 recommendation but still falls within the acceptable ranges from 0 to 1.0. Given that PLS-SEM emphasizes prediction above model fit, the outcome is regarded acceptable rather than ideal fit⁵⁸. The Goodness of fit, Model fit indicators, empirical results, and structural model are displayed in Tables 5, 6 and 7, and Fig. 3 respectively.

Discussion

The energy sector in Somalia has been operating for more than three decades under political instability and less regulated system, which led the emergence of a highly privatized and fragmented electricity markets that creates informal Electricity Service Providers (ESPs). Therefore, these private providers have indicated resilience in maintaining electricity access for Somali household. However, the higher tariffs and higher costs have created a limited consumer confidence and control over energy choices.

Globally, energy systems remain heavily dependent on fossil fuels, particularly lighting in homes, the energy we use daily is primarily powered by fossil fuels, the main drivers of climate change²³. Consequently, many nations, including Somalia, are over-reliant on fossil fuels. Burning fossil fuels releases large amounts of carbon dioxide, a greenhouse gas, into the air. In turn, greenhouse gases blanket the Earth and trap the sun’s heat, warming the planet with far-reaching consequences for people and the planet. However, renewable energy options are rising due to their ability to serve as a cleaner alternative to fossil fuels, which come with high production costs, contribute to environmental damage, and pose health risks to humans and animals²⁰.

This study investigates the factors that influence consumer intention and desire to utilize renewable energy, both directly and moderation effect. TPB was used as a theoretical framework for this study, with its original constructs attitude, subjective norms, and perceived behavioral control, focusing on consumer intention to use renewable energy, which has not been thoroughly studied in previous studies, as well as extending the theory to include two additional variables: belief about the costs of renewable energy utilization and environmental concern. These enhancements aim to provide a more thorough understanding of customer intention and willingness to adopt renewable energy in Somalia. The independent variables explained 42.8% of the variance in consumer WTP for renewable energy (R² = 0.428). The corrected R² of 0.402 verifies the model’s robustness, indicating that the predictors significantly help to understand consumer behavior regarding renewable energy uptake in Somalia.

The small coefficients in this study from 0.048 to 0.069 suggest moderate effects, even though their significance should not be determined just by magnitude. Even small but significant coefficients in PLS-SEM can reveal substantial relationships in complicated, multivariate systems with diffuse and interconnected effects.

Consumer intention to use renewable energy had the greatest impact on WTP for renewable energy (β = 0.069, p < 0.001, t = 3.001). As a result, this strong relationship between consumer intention and WTP renewable may not reflect only individual preferences but also growing expectations among households that Somalia energy sector is transitioning from fossil fuels to renewable and stable energy system. The National Electricity Authority (NEA) has been established recently in Somalia, which may increase the consumer confidence and strengthening the relationship between consumer intention and WTP renewable. This confirms the core assumption of the Theory of Planned Behavior, which states that behavioral intentions are the most proximal predictor of actual behavior³⁸. Similarly, the result aligns with^48,62,63, consumer intention has continuously been highlighted as an important connection between attitudes, norms, control perceptions, and actual behavior. Hence, Somali households have expressed a positive desire to adopt renewable energy are significantly more likely to pay for it. Second, Environmental concern has a significant impact on Somali households’ WTP for renewable energy (β = 0.068, p < 0.000, t = 4.102). The results are consistent with previous studies that has established the relevance of environmental protection in influencing pro-environmental behavior internationally^48,64,65, ,^66,67. Therefore, the findings suggest that environmental awareness and education may be particularly beneficial for boosting renewable energy use in Somalia and other comparable environments.

Third, perceived behavioral control has a positive and substantial relationship with WTP for renewable energy (β = 0.056, p < 0.000, t = 3.662). This result aligns with the Theory of Planned Behavior’s emphasis on the role of control perceptions in shaping behavior⁶⁸. This study also aligns with^48,63,69. Therefore, the perceived behavioral control indicates that it is more than just individual capability and reflects access to flexible and informal financial tools that enable households to manage energy payments. In addition, the highly accessibility of mobile money platforms including Sahal and EVC Plus allows households to make incremental payments by increasing their perceived control over renewable energy adoption. Forth, subjective Norms significantly increase WTP renewable energy (β = 0.064, p < 0.007, t = 2.448). Previous research has validated these conclusions^48,70,71. Hence, the result of collectivist social structure, family networks, community leaders, and local market players frequently influence the willingness to pay renewable energy in Somalia. These findings underline the need of using existing social structures to promote renewable energy as a socially acceptable and desired activity. Lastly, the study found a marginally significant positive relationship between attitude and WTP for renewable energy (β = 0.064, p < 0.066, t = 1.510). This supports the Theory of Planned Behavior, which states that attitudes impact behavioral intentions and subsequent behavior⁷². This also aligns with^48,63,64. Hence, positive attitudes towards renewable energy includes perception of its benefits, advantages, and attractiveness towards WTP renewable energy.

However, Belief About Renewable Energy Costs appears to have a negligible effect. Surprisingly, views regarding renewable energy prices had no significant impact on WTP (β = 0.058, p < 0.109, t = 1.231). However, this result was unexpected, as cost factors are commonly significantly effects the WTP renewable energy, particularly in resource-constrained environments as most prior studies suggested a negative influence of renewable energy costs on WTP for it⁴⁸. The insignificance of this particular result indicate energy-constrained environments which reliability and accessibility of electricity may be prioritized over cost considerations. Furthermore, the insignificance of cost beliefs might show that perceived benefits, environmental, social, and practical, outweigh cost considerations for those predisposed to adopt renewable energy. However, trust is critical in determining energy decisions in fragile states like Somalia, when formal institutions are scarce and customers rely significantly on familiar service providers and trusted financial platforms.

In terms of moderation, the findings show that consumer intention strongly regulates the correlations between environmental concern (H8), subjective norms (H9), attitude (H10), and WTP for renewable energy. These moderating effects show positive significant impact of environmental concern, societal norms, and attitudes on WTP renewable energy is moderated by consumer intention. These variables impact not just WTP for renewable energy, but also function in tandem with consumer goals. For example, a consumer who is concerned about the environment may be prepared to pay more for renewable energy if they have strong plans to use such technology. Similarly, positive social effects subjective norms may be most successful in increasing WTP for renewable energy among people who have already made tentative plans to utilise it. Consumer intention did not significantly moderate the relationship between Belief About Renewable Energy Costs (H7), perceived behavioural control (H11), and WTP for renewable energy. The lack of moderation in cost views is consistent with the discovery that cost beliefs did not significantly affect WTP directly. This trend suggests that cost factors are independent of intention strength, neither amplifying nor diminishing customers’ intent to use renewable energy. The lack of moderation in perceived behavioural control is potentially fascinating. It suggests that the empowering effect of perceived control on WTP operates consistently regardless of intention strength.

Theoretical and policy implication

This work provides numerous significant theoretical advances in understanding renewable energy uptake in underdeveloped countries. First, it establishes that TPB theory applicability and significance in the Somali context, where it has previously been underutilized. Attitudes, subjective norms, and perceived behavioral control have substantial influence on renewable energy use decisions across a variety of cultural and economic contexts. Second, the study broadens TPB Theory by including Belief about Renewable Energy Cost and environmental concern as predictors, suggesting that environmental values play an important role beyond typical TPB concepts. Given the considerable impact of environmental concern, models of pro-environmental behaviour in energy contexts should consistently incorporate environmental values and concerns as crucial predictors. Third, investigating moderating effects yields detailed insights into how many factors interact to impact WTP for renewables. The discovery that consumer intention modifies certain associations but not others implies that the paths from psychological antecedents to behaviour are complicated and conditional, necessitating more sophisticated theoretical models that account for interaction effects. Somalia confronts substantial energy access issues, with little grid infrastructure and a high dependence on diesel generators and traditional biomass for electricity⁷³. In this context, renewable energy particularly solar power represents not merely an environmental choice but a practical solution to energy access problems.

The results of this study yield significant insights for policymakers, energy strategists, and advocates of renewable energy. Initially, it is imperative that policymakers establish a tiered licensing framework that facilitates the gradual integration of renewable energy technologies by existing private energy service providers (ESPs), while simultaneously providing them with the necessary technical and regulatory assistance. Furthermore, financial institutions alongside telecommunications enterprises should engage in collaborative efforts to devise mobile-based “pay-as-you-go” financing models for renewable energy, specifically designed for households lacking access to formal credit systems. Moreover, the establishment of community-oriented energy cooperatives, spearheaded by local leaders and market associations, has the potential to bolster trust, decrease costs, and expedite the adoption of renewable energy solutions. Additionally, the implementation of direct subsidies, tax incentives, or rebate schemes is recommended to alleviate the initial financial burdens associated with renewable energy installations. In conclusion, these findings elucidate that the WTP renewable energy in Somalia is influenced not solely by individual perspectives but also by the levels of institutional trust, informal market frameworks, and socio-cultural factors.

Conclusion

This study discusses the factors influencing WTP renewable energy in Somalia. The study extends the Theory of Planned Behavior by incorporating two more variables: Beliefs about the costs of renewable energy and environmental concern. The study also examines the moderating role of consumer intention. The result indicates that attitudes, subjective norms, perceived behavioral control, consumer intention, and environmental concern have significant positive impact on WTP renewable energy in Somali households. While Beliefs about the costs of renewable energy do not directly influence on WTP renewable energy in Somalia. Moreover, the results also highlight that the consumer intention moderates between environmental concerns, social norms, and attitudes on WTP renewable energy. However, policymakers and energy stakeholders in Somalia, including electricity providers, should prioritize strategies that increase positive attitudes and foster supportive social norms, as these efforts, when combined with strong consumer intention, are critical accelerating the transition to sustainable energy in Somalia.

The limitations include, the cross-sectional design precludes causal conclusions because all variables were assessed at a single time point. Second, the analysis is based on self-reported WTP for renewable energy, rather than actual adoption behavior or payment statistics. Third, the analysis explains 42.8% of the variation in WTP for renewable energy, suggesting that additional key aspects remain to be investigated. Future studies should examine additional determinants, such as economic factors like income and access to funding. Finally, because the study was conducted in Somalia, generalizability to other contexts should be assessed through replication studies across nations and cultural contexts. While the findings are likely applicable to other developing countries facing comparable energy issues, local variables may change the relative impact of several predictors.

Abbreviations

RE

Renewable energy

WTP

Willingness to pay

PBC

Perceived behavioral control

ECN

Environmental concern

TPB

Theory of planned behavior

CR

Composite reliability

AVE

Average variance extracted

VIF

Variance inflation factor

SEM

Structural equation modeling

PLS-SEM

Partial least squares structural equation modeling

HTMT

Heterotrait–monotrait ratio

SRMR

Standardized root mean square residual

R²

Coefficient of determination

Author contributions

Galad Mohamed Barre wrote the introduction, literature review, econometric methodology, data collection, analysis, interpretation of results, discussion section, manuscript writing, and Ahmed Hassan Mohamud reviewed the manuscript.

Funding

This research was supported by Simad University.

Data availability

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

All methods were conducted according to relevant guidelines and regulations. The research protocol, survey tools, and consent process were approved by the Ethics Committee of the Centre for Research and Development at SIMAD University (Approval Number: EC000138). Informed consent was obtained from all respondents before data collection. Participation was voluntary, and all respondents were assured of the confidentiality and anonymity of their responses.