Influence of policies towards building retrofitting for energy efficiency in Zimbabwe using structural equation model

Globally, the built environment sector generates about 39% of energy and process-connected carbon emissions [International Energy Agency (IEA), 2019]. Sun et al. (2021) reported that between 1990 and 2020, energy consumption grew by 42.5%; likewise, carbon emissions increased by 32.4%. In 2018, Prabatha et al. (2020) reported that 11% of emissions were connected to construction and building materials production. Also, building and construction installations and accessories generate some embodied energy in construction projects (Tokede et al., 2023). Embodied energy is described as energy connected with materials, conveyance and mining maintenance (de Oliveira Fernandes et al., 2021). The work of Costa-Carrapiço et al. (2020), Dzobo et al. (2020), Jo et al. (2022) and Geh et al. (2024) advocates that retrofitting of buildings can improve energy efficiency and sustainability, especially in developing countries such as in Southern Africa. Hepner (1994), Oladokun and Aigbavboa (2018), Costa-Carrapiço et al. (2020), Jayarathne et al. (2024) and Geh et al. (2024) also emphasised the importance of retrofitting buildings to improve energy efficiency. Hence, it becomes germane for stakeholders involved in the energy industry in Zimbabwe to embrace energy efficiency through building retrofitting to enhance sustainability and save future costs attributed to repairing buildings and fuel poverty (Oladokun and Aigbavboa, 2018).

Having an awareness of policy formulations in Zimbabwe and their impact is essential, especially when it comes to suggestions for policy and concept revisions to ensure that retrofit policies in the construction industry are implemented (Ovald, 2024). Take, for instance, which identifies how land acquisition by local Zimbabweans from White farmers stands in the way of energy efficiency projects. In addition, it also identified how unskilled personnel in energy efficiency was also a challenge which slowed down energy efficiency adoption in the country (National Renewable Energy Policy (2019). The National Renewable Energy Policy (2019) is a cornerstone of Zimbabwe’s efforts to promote sustainable energy practices. This study critically evaluates the policy’s framework and its implementation, providing a robust basis for the SEM analysis. The integration of Policy Implementation Theory further enhances the study’s theoretical foundation by examining how policy design, implementation and outcomes interact in the context of retrofitting. The context of Zimbabwe makes us increasingly aware of contextual factors, that prohibit the uptake of energy efficiency in building retrofit projects, which in the main are to do with financial outlay involved (Okorafor, 2019). Hence, there is a pressing need to retrofit existing buildings to save money and to prolong the lifecycle of buildings, thus helping ease overcrowding in Zimbabwe’s urban areas. The objective of this study was to find out how policy action affects the retrofitting of buildings for energy efficiency in Zimbabwe. Preliminary findings show a large existing stock without energy efficiency principles in Southern Africa and might face complications in upgrading (Kanda et al., 2022; Alaloul et al., 2020). This is because energy efficiency is not in high demand in low-income housing such as that found in most African countries (Okorafor, 2019).

For this study, defined retrofitting as the installation of an individual or multiple energy efficiency measures to an existing building. Steskens et al. (2015) are concerned that there is no motivation to retrofit buildings. This is pronounced in developing countries’ stakeholders. There is a call and incentive to embrace retrofit projects given the focus on sustainable development goals (SDGs), as well as the need to combat the negative impact of climate change (Steskens et al., 2015). In addition, Zimbabwe has faced floods that have led to the loss of lives and property. Hence, as outlined earlier, the focus of the study is to appraise policies that influence the retrofitting of buildings for energy efficiency in the Zimbabwe construction industry. The use of the structural equation model will be presented as an innovative way in which to approach policy studies in sub-Saharan Africa. This will depart from the Multiple Streams Frameworks, which identify the different streams that affect policy action (Steskens et al., 2015), to some extent, this study adds to this theory by advocating the use of inferential statistics in foreground policy studies. Hence this study will appraise how a retrofit model identifies the main influencing policies as a possible approach to improving a building’s energy efficiency. As identified by Chigudu (2015), the biggest challenge for most African societies is not making policies but implementing the policies that they have. Hence, this is one of the gaps (locational) this study will fill. In Southern Africa context, the research gap, also coheres with the work of Dzobo et al. (2020), who whilst exploring the ways in which energy efficiency in the Zimbabwe’s manufacturing sector can be improved, identified inadequate support from senior management, insufficient data on energy efficiency programmes, inadequate coordination between firm divisions, risk of production disruption and lax structured electricity tariffs as hindrances to energy efficiency.

These issues highlight the complexity of policy action, which this study has sought to simplify using structural equation modelling. The conceptual framework underpinning this study is that of the Multiple Streams Framework (MSF), which is based on the idea that policy action results from a convergence of different issues, such as problem stream, policies and politics stream. If these converge that is when policies are implemented (Ovald, 2024). Thus, this study extends the application of SEM by integrating it with the MSF to analyse policy influences in a developing country context, specifically Zimbabwe. Unlike previous studies that have primarily focused on developed countries, this research provides unique insights into the challenges and opportunities in a developing country setting using Zimbabwe as a case study. It is when these streams converge that policy action takes place; this study sought to use MSF to try and understand how policy action occurs in Zimbabwe whereas at the same time using the structural equation model as an approach to try and understand the variables and alternatives that are there to facilitate decision making (Ovald, 2024). Therefore, in this study, the MSF is being used as an aid to decision-making, and the use of an SEM, which will be explained later support the selection of variables that policy entrepreneurs have to consider if they are to take action to retrofit existing buildings in Zimbabwe. However, SEM ensures that the measurement of the latent constructs and indicators are adequate and credible to meet the rules of quantitative studies (Dirgiatmo, 2023).

Retrofitting provides an opportunity to improve structures’ energy and emissions performance, especially buildings (Tokede et al., 2018, 2023; Mapfumo et al., 2024). Retrofitting alters structures (buildings) to improve their sustainability in terms of carbon emissions reduction (Oladokun and Aigbavboa, 2018). In this context, sustainability provides an opportunity for upcoming climate conditions, reduces environmental emissions and optimises economic building benefits (Jo et al., 2022). Menassa (2011) asserted that retrofitting is a considerable intervention that improves performance and enhances building certainty usage over a defined period. As noted by Tokede et al. (2018) policies that support retrofit for energy efficiency encourage building energy efficiency and sustainability. Based on the studies of Dutch residential buildings, the work of Berg and Fuglseth (2018) and Tokede et al. (2023) makes us aware of the potential benefit of retrofitting for energy efficiency, as they found that it could result in a 60% decrease in the overall environmental impact produced over the operational life cycle stage. A different study by Hasik et al. (2019) found that up to 75% less environmental impact could be achieved, with finishes and glazing accountable for a large amount of carbon emissions. There are many ways in which to look at retrofitting, and it is worth noting that in Zimbabwe, there are challenges to do with the absence of electrical power, such that it is not much about conserving energy but about how people survive with no energy. This perhaps broadens the discussion on energy efficiency, especially if one takes note of the view of Dixon et al. (2014), who highlighted that retrofitting buildings does include investigating water and waste efficiency opportunities.

In South Africa, Oguntona et al. (2019) found low consumer appeal, high upfront cost, occupant resistance, high investment costs and low income as the top five barriers facing building retrofit projects. However, the benefits of retrofitting buildings cannot be over-emphasised. It can drastically reduce carbon footprint, green gas emissions and long-term cost of energy consumption. Amoah and Smith (2024) corroborated Oguntona et al.’s (2019) findings and opined that understanding the barriers will enhance informed decisions to adopt green retrofitting principles in South Africa. Magaisa et al. (2022) reviewed how South African Government buildings can be retrofitted to improve safety from fire. In Zimbabwe, there have also been issues regarding the disposal of waste, and finding solutions to these challenges, through energy efficiency policies can go a long way in improving the environmental conditions and atmosphere. In the UK, Tokede et al. (2018b) found that retrofitting heritage office buildings for energy efficiency will enhance cost savings and mitigate the construction industry’s impact on climate change and global warming. Hence, this means that Zimbabwe’s old historic buildings that were constructed in the colonial time in the 60 s can be retrofitted for energy efficiency and cost savings with the right policies (Kanda et al., 2022). This may become a red flag if the relevant stakeholders do not give urgent attention to existing buildings and how they can save the Zimbabwean government money, and the health and wellbeing of the nation.

Ferrando et al. (2020) studied thermal comfort in Iran and found that houses were not examined to confirm compliance with the state’s rules and regulations. This study is also relevant when trying to understand Zimbabwe’s context, as noted by Chigudu (2015) who argued that the failure to implement well laid down policies was the downfall of Zimbabwe’s policy framework in different fields including construction and the energy sector. Foruzanmehr’s study, as cited in Ferrando et al. (2020, p. 12) found, […] .in hot climates, without electro-mechanical cooling systems, most 20th century buildings are not suitable, even for present climatic conditions […] This implies that context awareness is critical to traditional thinking methods. This has been upgraded in many developed countries. Zimbabwe’s climate is hot, and the perspectives by Foruzanmehr have to be considered, because some retrofit measures applied in European countries might not be applicable. Ferrando et al. (2020) corroborated. Konya’s study, as cited in Ferrando et al. (2020, p. 7), said, Modern dwellings have been designed largely to keep natural phenomena outside and to separate conditions indoors from outdoors as much as possible. Eriksson et al. (2020) found that historic structure preservation in Sweden can influence the energy-saving potential. It implies that historic structure preservation is a form of policy influencing retrofitting for energy efficiency.

The United Nations 2030 Agenda has reawakened member countries, including Zimbabwe, to promote policies tailored towards improving energy efficient through building retrofitting. Zimbabwe’s Government goal is to improve the achievement of SDGs related to energy efficiency and affordability for the masses, especially the social houses (Geh et al., 2024). Policies regarding energy-efficient buildings are very important when focusing on how construction projects are managed in Zimbabwe (Charis et al., 2019). This is because policies and how they are followed can impact future generations (Hugo et al., 2021). In Zimbabwe, some scholars highlighted how policies are intertwined with politics, creating a situation in which each political party will tend to promote certain policies (Chipango, 2020; Dube and Nhamo, 2020; Ncube et al., 2021). This also makes the perspectives of the multiple streams framework interesting, because it highlights the importance of the policy stream in politics (Kanda et al., 2022; Howlett, 2019). Hence, this study brings to our attention the need to be aware of a multitude of influences surrounding policy studies; that the influence of politics in policy formulations impacts decision-making, as the government becomes persuaded by a majority vote. In addition, getting occupants post occupancy feedback and evaluations on building performance can be a challenge, especially if questions are to be asked on their thermal comfort in existing old buildings. The challenges of implementing well-laid down policies continue to cripple the government policy-making effectiveness (Kanda et al., 2022; Chigudu, 2015). There is a need to re-imagine what policies would work better, considering the alternatives and issues being discussed here (Howlett, 2019). Not forgetting how financial hardships add to the challenge for the Zimbabwean Government to account for how policies benefit the economy (Kativu and Oskarsson, 2021; Chirisa et al., 2021). Also, the lack of competency of construction practitioners regarding the expected role in enhancing energy efficiency has not helped matters (Moyo, 2020). Lee (2021) argued the need for policies to be tailored towards addressing complex situations. The energy efficiency policies are not exempted, especially now that SDGs are fast approaching (Ahmed et al., 2021; Fernandes et al., 2021). The research method will seek to provide alternative ways to engage with policy studies in an African context, using inferential statistics based on real-life scenarios and problems.

The main research’s objective was to identify and assess policies influencing building retrofitting for energy efficiency in Zimbabwe. The choice of Zimbabwe was based on increasing literature that has been centred on energy use in buildings in the construction industry Zimbabwe and a renewed focus on retrofitting of buildings (Dzobo et al., 2020). In addition, the need for green building codes to improve sustainability in buildings, as well as policy changes which have been aimed at improving energy efficiency and creating smart cities have also been the driver for this research (Chirisa et al., 2021) Given the issue of overcrowding in urban areas in Zimbabwe, it became critical to also test some of the frameworks, surrounding policy studies. By assessing how policy formulations influence the retrofitting of buildings for energy efficiency in Zimbabwe, this research intends to discover insights that can aid other African countries in making effective policies in the light of SDGs as well as the transition to net zero, which has also been the focus for several African countries (Geh et al., 2024). The study respondents were mainly key stakeholders involved in energy retrofits, which targeted the Ministry of Energy and Power Development (MoPD), Zimbabwe. This study adopted a pragmatism approach (Oke et al., 2024) and employed a mixed method approach. This study reports on quantitative results using a questionnaire survey method to proffer answers to the study’s main questions. This methodological choice agreed with Aghimien et al. (2021), who conducted a similar study in the same area using quantitative methods.

The study conducted a pilot study to ensure the efficiency and precision of the data collection instrument. It aligns with Quiles et al. (2019) and Ebekozien et al. (2019), who adopted the same approach in their studies and emphasised the need for a pilot study to refine and validate the survey instruments before the main distribution. The researchers slightly modified the data collection instrument based on the feedback from the pilot study to enhance the reliability of the results. The questionnaire comprised two sections (background and the section that ranked the influencing policies). The study used different communication mediums to administer the questionnaires to the respondents to increase the response rate. Google form, email and face-to-face were selected as methods of data collection during questionnaire administration to ease accessibility. Fifty-six usable questionnaires were retrieved and adopted for analysis from the 120 respondents’ sample size through cluster sampling, which targeted the Ministry of Energy and Power Development (MoPD and other stakeholders. The stakeholders include academia, engineering, commerce and construction, and achieved a response rate of 46%. The researchers accepted the study’s response rate because it was higher than the norm of 20–30% benchmark with questionnaire surveys of the construction industry (Akintoye and Fitzgerald, 2000). Moser and Kalton (1971) found that the response rate of research conducted in the construction industry is usually within 30–40% benchmark, and a rate lower than the range would be considered biased. The study adopted a five-point Likert scale to measure the level of agreement of the identified influencing policies, with 5 = strongly agree, 4 = agree, 3 = neutral, 2 = disagree, 1 = strongly disagree.

The study employed Statistical Package for the Social Sciences (SPSS) to analyse collected data. The researchers conducted a reliability evaluation of the study’s instrument through the calculation of Cronbach’s alpha coefficient (Oke et al., 2024). This mechanism offered an empirical perception of the internal uniformity of the questionnaire dimensions/items. An adequate Cronbach’s alpha value shows a strong level of reliability (Bujang et al., 2018; Oke et al., 2024; Ebekozien et al., 2025). It implies that the study’s instrument consistently measured the envisioned variables. In this research, 20 identified influencing policies displayed Cronbach’s alpha values ranging between 0.826 and 0.714. This signifies a high level of acceptance and reliability of the study’s instrument because the values are above the 0.7 threshold suggested by Tseng et al. (2006). The study examined the first phase of the questionnaire using descriptive statistical measures only. Also, the data normality was evaluated using the Shapiro–Wilk test (Hanusz et al., 2016). Mean score ranking and standard deviation were used to rank identified influencing policies regarding the level of agreement.

The exploratory factor analysis (EFA) was adopted to discover items and patterns within the assembled data (Luo et al., 2019). Kaiser–Meyer–Olkin (KMO) test and Bartlett’s sphericity test were undertaken to check the collected data’s appropriateness for factor analysis (FA). The KMO test was utilised to evaluate the sampling adequacy by measuring the proportion of variance in the constructs. Bartlett’s test of sphericity assesses whether the constructs in the data set are linked to a sufficient degree to continue with factor analysis (Oke et al., 2024). If the p-value linked with Bartlett’s test is statistically significant (below 0.05), it is recommended that the connections between constructs are significant for factor analysis (Aliu et al., 2024a, 2024b; Oke et al., 2024). Factors extracted from the use of FA had to be tested to ensure that they were valid, and this was done by using Confirmatory Factor Analysis (CFA), which was used to confirm the relevance of the factors. Also, a model was developed using the structural equation model (SEM) technique to aid decision making among key stakeholders and government officials in Zimbabwe. The SEM was used to determine the factors that influence policies. Its use also helped to determine the factors underpining policy formulations regarding energy efficiency in existing buildings in Zimbabwe. Despite the uncommon use of SEM in construction more generally around the globe, this method would be innovative and useful in the Zimbabwean context, given that the relationship between policy formulation and implementation is non-linear (Dirgiatmo, 2023). SEM was chosen over alternative methodologies, such as grounded theory or qualitative comparative analysis, due to its ability to model complex relationships between multiple variables. A critical review of empirical studies in the field, such as [cite specific studies reviewed in the literature review section (note that this is yet to be done in this study)], highlights the strengths of SEM in analysing policy influences. Constructs were selected based on their relevance to the National Renewable Energy Policy (2019) and their alignment with the MSF (another outcome from a critical literature review). The moderate response rate of 60% is discussed in the limitations section, and its implications for generalizability are addressed. Dirgiatmo (2023) emphasised that SEM is most appropriate because it can be used to determine the discriminant validity violation and used for the Heterotrait–Monotrait ratio of correlation (HTMT). The HTMT can detect discriminant validity violation (Guenther et al., 2023). Guenther et al. (2023) and Rahman et al. (2024) opined that SEM is used to theoretically establish a model with constructs and aligns with this study’s aim through the measurement model. This methodological approach was missing in the study’s geographical extant literature.

The respondents’ background information includes their years of experience, membership status and professional and academic qualifications. The respondents’ academic and professional experience shows a good representation across those with the least years of experience (ten years or less). These insights from the respondents give credibility to the findings because most of them had experience of retrofitting buildings for energy efficiency in Zimbabwe. Findings agree with Aliu et al. (2024). Table 1 shows the summarised responses relating to the influence of policies towards Zimbabwe’s building retrofitting for energy efficiency. The respondents indicate the degree to which each statement is related to determining the influence of policies towards Zimbabwe on a Likert scale. Table 1 shows that the mean scores of 20 variables are above the midpoint of 3.00. This indicates that the respondents, in general, agree/strongly agree that the items being measured are related to the influence of policies towards Zimbabwe’s building retrofitting for energy efficiency. The mean score and standard deviation were calculated for each construct to determine the frequency and the number of individuals who agreed with the statement (variable). The results show that renewable energy policy is relevant, especially when retrofitting buildings for EE in Zimbabwe, and was thus ranked highest with a mean score of 4.66 (standard deviation = 0.92). The second-ranked factor had to do with energy regulations policies that can save the overall cost of projects (energy efficiency is cost saving) with a mean score of 4.59 (standard deviation = 0.91). Contractors were identified as a barrier to EE in retrofit projects, thus influencing how policies are made, and ranked least with a mean score of 3.52 (standard deviation = 1.06).

This sub-section presents EFA and CFA results. First, EFA was undertaken, followed by CFA, to confirm the results of EFA and to take note of the key influence of policies towards building retrofitting for energy efficiency in Zimbabwe. EFA was used to determine the constructs (factors) that influence how policies towards building retrofitting for energy efficiency in Zimbabwe. First, the researchers evaluated all 20 items (influencing factors). Table 2 shows how 12 factors were extracted based on high loading values, which meant that they were influential on how policies were made for energy efficiency in Zimbabwe. The factors were determined using Kaiser’s eigenvalue cut-off point, which requires that this value is greater than, or equal to, 1. Components 1, 2 and 3 had eigenvalues of 4.957, 1.646 and 1.452, respectively, and explained 41.305%, 13.715% and 12.099%, respectively, of the variance. These three selected factors collectively accounted for 67.119% of the total variance observed.

Table 3 presents factor loading (above 0.4), communalities (CM) Cronbach’s alpha, average variance extracted (AVE) and composite reliability (CR) on key policies impacting building retrofitting for energy efficiency in Zimbabwe. The first component, the utilization of enabling policies, had factor loading ranging from 0.825 to 0.504. The observed CM for the items in the first factor ranged from 0.330–0.748, indicating a satisfactory explanation of the extracted factor. Cronbach’s alpha yielded an observed value of 0.826, AVE of 0.673 and CR of 0.453. This affirms the reliability, validity and correlation consistency for the items under the first factor, which is about enabling policies that influence building retrofitting in Zimbabwe. The second factor pertains to raw material specification, with factor loadings ranging from 0.833–0.473. The study observed CM for these items ranged from 0.275–0.779, indicating an adequate explanation for the extracted factor. Cronbach’s alpha yielded an observed value of 0.842; AVE had a value of 0.704, and lastly, CR of 0.496. This satisfied the reliability, validity and correlation consistency observed for items under the specification of raw materials for building energy efficiency retrofitting in Zimbabwe. The final factor examines guidance provided to stakeholders in building energy efficiency for the items with factor loading ranging from 0.759 to 0.531. CM of the same items ranged from 0.421 to 0.675, indicating adequate explanation of the extracted factor. Cronbach’s alpha reliability was 0.714, AVE’s value of 0.656 was observed for items related to guidance for energy efficiency stakeholders. The results show reliability, validity and correlation consistency satisfied policies that influenced building energy efficiency retrofitting in Zimbabwe.

The researchers conducted CFA to confirm the results of factor analysis, and model fit indices helped in coming up with a SEM which was subjected to validation before being adopted. The study used the robust maximum likelihood (RML) estimation because the data did not conform to normality. The SEM was employed to confirm the significance of the factors extracted from EFA. As shown in Table 4, the initial model yielded a poor result. For instance, GFI, RMSEA and NFI did not attain the threshold values. Hence, backward selection (removing items with the lowest loading) was undertaken until optimal measures were attained, which resulted in the final model. For the first factor, enabling policies, the item “use of sunlight and conventional ventilation” was removed. In contrast, the item “different interpretation of a business model” was removed from the second factor, the specification of raw materials. This was also the case for the third factor, guidance to stakeholders involved in the energy efficiency of buildings, “government agents are a barrier,” was removed from the model.

Discriminant validity of the factors attained in the final model of the SEM for policies influencing the retrofitting of buildings for energy efficiency in Zimbabwe was undertaken, as presented in Table 5. Table 5 presents the HTMT to assess the discriminant validity of three factors: enabling policy, specification of raw materials and guidance to stakeholders involved in the energy efficiency of buildings. The HTMT was performed to check if the three factors were related to policy, and the results suggest that the values were significant, which indicated that they were related to policy influences. Therefore, it suggests that the developed SEM was acceptable and could highlight the main themes to consider before implementing policy. Thus, confirming the distinctiveness between the factors.

Figure 1 reveals the final SEM after removing variables with low factor loadings. CFA showed that the first factor focuses on enabling policies supporting retrofitting buildings in Zimbabwe. One novelty contribution from the study is the cluster of influence policies into three main clusters. This includes enabling policies, specification of raw materials and guidance to stakeholders involved in the energy efficiency of buildings. For the enabling policies, findings group the use of solar energy with the need to focus on energy regulations, climate change policy and energy policy as essential. This means that energy regulations that contractors and other stakeholders can save costs for the government of Zimbabwe, as well as for construction companies because it will help to ensure that retrofitting efforts are supported. This model could be adapted by other developing countries with similar challenges regarding building retrofitting. This means that the government of Zimbabwe must consider how cost savings can be made when creating energy efficiency policies. The second factor, the specification of raw materials, which to some extent is also captured in Zimbabwe’s biofuel policy, is essential regarding energy regulation. In addition, the National Energy Policy of Zimbabwe is critical regarding energy regulation, and so is the use of recycled material such as plant waste and use of grey water from natural rain. The results, as shown by the CFA, are that the third factor, which has to do with guidance to stakeholders involved in the energy efficiency of buildings, designers and contractors, is influential in ensuring that energy-efficient policies in Zimbabwe are adopted in the construction industry. Hence, policy development must consider the views of stakeholders and what can be done to get their full support. The questionnaire was developed based on a literature review on energy retrofitting buildings in Zimbabwe, and the responses from experts in the field of energy retrofits in Zimbabwe. Hence, the SEM developed highlights the situation regarding retrofitting for energy efficiency in Zimbabwe using quantitative research design method.

Table 6 reveals the SEM that helped to test the effectiveness of policies enacted regarding retrofitting buildings for energy efficiency in Zimbabwe; this includes a number of indices such as the standardized coefficients, p-values, CR and coefficient of determination (CD). These findings suggest that the standardised coefficients exceeded 0.5, which means that each variable explained more than 50% of the variance in the model, indicating robust constructs. The p-values are less than a 5% level of significance, thus portraying the significant influence of the items on the specific factors. In addition, the CR values surpassed 0.40, indicating strong internal consistency and reliability within the SEM. Further, the CD showed the variability of each item in relation to a particular factor. On the third factor of guidance to stakeholders involved in the energy efficiency of buildings and contractors’ areas, a barrier to energy regulation had the highest CD of 0.866, suggesting it explains approximately 86.6% of the total variance. Conversely, enabling policies had the lowest explained variance of energy regulation’s cost-saving aspect, with a value of 0.307, indicating a weaker relationship between the item and the factor. This reinforced the challenge of cost when it comes to adopting policies in Zimbabwe.

The findings highlight the need to tailor policies towards improving raw materials to construct buildings that are energy efficient. Other components include paying attention to how projects are funded and by who (finance) and the human skills of construction workers. What is apparent is that raw materials are crucial and may influence building retrofitting for energy efficiency in Zimbabwe. Hence, energy policies should focus on specifying materials that can be used on energy efficient buildings. In addition, there will be need for building codes that are based on specific raw materials. One unexpected finding was the low ranking of contractors as barriers to retrofitting. This suggests that whereas contractors play a crucial role, other factors, such as policy support and stakeholder guidance, are more significant. Therefore, the government of Zimbabwe, through its Ministry of Local Government can also seek to find ways in which they can subsidize some of the construction raw materials to ensure that energy-efficient buildings can become a reality for most people (Moyo, 2014). Findings agree with Kılkıs et al. (2023) and Geh et al. (2024). Geh et al. (2024), who acknowledged the contributions of retrofitting buildings for energy efficiency, because it is also tied to SDGs. Stakeholder-specific perspectives reveal that policymakers, industry experts and community members have distinct views on the challenges and opportunities. These findings contribute to the SDGs by highlighting the importance of enabling policies and stakeholder engagement in promoting energy efficiency. The SDGs are associated with how energy is used, and include SDG 6 (ensure availability and sustainable management of water and sanitation for all), SDG 7 (ensure access to affordable, reliable, sustainable and modern energy for all), SDG 11 (make cities and human settlements inclusive, safe, resilient and sustainable) and SDG 13 (take urgent action to combat climate change and its impacts) (UN, 2022). It implies that retrofitting buildings for energy efficiency via policies would improve access to affordable drinking water management, energy, housing and sustainable climate. Despite these benefits, the application is low and more pronounced in many developing countries, including Zimbabwe.

Findings emphasise enabling policies that can birth raw materials for energy efficiency and inclusiveness of all stakeholders, especially in residential buildings. In this study, the use of cluster sampling means that key stakeholders involved in energy retrofitting in buildings are considered through the Ministry of Energy and Power Development in Zimbabwe (MoPD). This has become pertinent because of the increasing human activities in constructing and operating buildings and other related areas, leading to Greenhouse Gas (GHG). Findings agree with the work of Baker (2016), Tanyanyiwa and Juba (2018) and Geh et al. (2024), regarding the need to ensure that contextual factors are considered that the reviewed literature noted that the unprecedented increase in GHG levels is the cause of climate change and hence this calls for policy action to mitigate the “red flag warning” (UN, 2022). Their findings corroborated the report by Climate Change 2022 that in 2019, 21% of global GHG emissions came from buildings (IPCC, 2022). UNEP (2021) suggested that to mitigate emissions, especially in buildings, the need for stakeholders to significantly reduce energy demand, acceleration of electrification through energy regulation and selection of lower carbon building materials and emissions from construction processes cannot be over-emphasised.

Findings show that the building industry is a heavy consumer of raw materials; thus, Goal 12, which targets sustainable consumption and production patterns, is directly related. The industry tasks are known to meaningfully contribute to carbon emissions, especially old, constructed buildings. Findings agree with May and Griffiths (2015), Che Husin et al. (2019) and Geh et al. (2024). Geh et al. (2024) recommended that constructed buildings can be improved by retrofitting them to improve energy efficiency and performance. May and Griffiths (2015) and Che Husin et al. (2019) suggested that including green building technologies in retrofitting is paramount to enhancing building efficiency and performance regarding energy supply. Some green building features include energy regulation technologies, upgrading, or replacing air-conditioning to mitigate energy consumption, installing building management systems and using LED lights. Findings agree with Chivhenge et al. (2023) and Sithole et al. (2023). Chivhenge et al. (2023) identified steps being taken by Zimbabwe’s Government to mitigate the emissions targets on or by 2030. This includes encouraging the people to use electric cars, renewable energy, fuel blending for decarbonisation and the establishment of national climate financing mechanisms. Achieving these targets has been challenging because of policy conflicts. Again, it validates the study that policy could influence retrofitting Zimbabwe’s buildings for energy efficiency. Sithole et al. (2023) asserted that Zimbabwe’s Government targets a 40% reduction in greenhouse emissions by 2030. To achieve this target, 17-point mitigation action plans have been instituted. They emphasise that the mitigation should be all-inclusive, and specifically, the forestry sector is crucial, followed by the energy sector. These mitigations could enhance substantial local development benefits, including the use of locally available raw materials for sustainable energy use and improving public health in Zimbabwe (Maphosa et al., 2020). The role of the local government and collaborations with key stakeholders are critical in mitigating greenhouse emissions. Findings agree with Tinarwo et al. (2023), who affirmed that government interventions can improve building energy performance with the support of other partnerships with companies, especially construction companies of different sizes that are constructing buildings for energy efficiency. It enhances interventions based on best practice solutions embedded in suitable enabling situational realities. Tinarwo et al. (2023) findings corroborated Tokede et al. (2018b), who suggested incentivising building retrofitting, especially heritage buildings, to mitigate the negative impact of climate change and global warming. Findings reveal that energy efficiency performance is a global issue and requires policies to enable the Zimbabwean construction and energy sector to be strategic and to focus on how they can empower the next generation regarding energy efficiency and sustainability. This aligns with Khan et al. (2021), who affirmed that the European Union targets 80% reduction of carbon emissions before 2050.

This research assessed the influence of policies towards retrofitting Zimbabwe’s buildings for energy efficiency and developed a structural equation model that highlighted factors that must be considered to influence the energy efficiency policies in Zimbabwe positively. This study also demonstrates that policies and regulations significantly influence the retrofitting of buildings for energy efficiency. It also emphasises the need to find alternative and affordable methods to ensure energy gains and effectiveness are accessible to all. Through considering alternatives and using MSF to try and understand the dynamics of policy formulations, the study also highlighted some of the competing challenges that influence policy. Focusing on the benefits that can be harvested from sound policies, findings show that besides improving the water management system, sustainability and energy efficiency, retrofitting for energy efficiency also contributes to SDGs (6, 7, 11, 12 and 13). Besides the developed model, the three new themes from the clustered main policy action via the structural equation model form part of the study’s implication. The findings suggest a need for the government of Zimbabwe (policy entrepreneurs) and the community at large to think of green buildings through the engagement of stakeholders in building retrofitting for energy efficiency and creating sustainable communities. Stakeholder’s views and opinions are pertinent, especially when retrofitting buildings for energy efficiency. The study concludes that energy-efficient buildings will improve how energy is used and conserved, thus improving the nation’s health and well-being through policies and achieving SDGs related to energy consumption. The individual impact of this is that it reduces fuel poverty, as they can work or live in buildings that enhance their health and wellbeing, thus also prolonging their life expectancy. This study has shown that policies and regulations can influence the retrofitting of buildings for energy efficiency. However, it has also noted alternative and affordable ways to make energy gains and effectiveness accessible to all.

The study advocates areas that need to be considered when it comes to policy formulations because they influence how the retrofitting of buildings for energy efficiency takes place in Zimbabwe for the community. Also, policies relating to energy efficiency tend to focus on mitigating gas emissions, especially carbon emissions. Still, this study has highlighted the complex nature of policy formulations and adds to the literature on policy studies by introducing SEM as a tool by which policy makers can become more aware of latent variables that affect policy action. Thus, the result of the study identifies how complex policy studies are, and how non-linear this process is and hence policymakers and other key stakeholders concerned with energy efficiency should also consider green building perspectives and the use of alternative building materials technologies to retrofit buildings. Such thinking, especially in an African context, will mitigate the reliance on conventional cooking methods (firewood approach) and make energy efficient measures accessible and affordable to the public. These two elements are germane for the low-income earners in social housing. Policies applicability in the African context is germane to enhance embracing and sustainability that will yield all-inclusiveness. As part of future studies, the model and the emerged constructs should be further validated through a mixed-methods approach with wider coverage, including users/residents.

The authors special thanks to the respondents for providing scholarly contributions to enhance the findings of this paper. Also, the authors appreciate the comments, suggestions, and recommendations provided by the anonymous reviewers, which helped hone and strengthen the quality of this manuscript during the blind peer-review process.

Funding: Financial support from the Department of Construction Management, Nelson Mandela University, South Africa.