Corporate social responsibility pyramid in Ethiopia: A mixed study on approaches and practices

This paper examines CSR approaches and practices of local firms in Ethiopia, where the state is still a key player in business life. Moreover, it assesses at which stage Ethiopian CSR is. This mixed-method study includes quantitative analysis through SPSS and qualitative content analysis. A revised CSR practices scale has been conducted on owners/managers and employees of 100 local companies in Ethiopia. Besides, the companies' official websites have been thoroughly reviewed using five variables to understand their CSR approaches and practices better. The findings reveal that Ethiopian companies practice CSR though it is at the stage of fragmented CSR and the priorities differ from the global trend. In general, most respondent companies have excellent legal practices and CSR practices toward employees and customers. However, philanthropy, social sustainability, and management of CSR are remarkably at a mediocre level. Content analysis shows similarity with the results of quantitative study, to the exclusion of philanthropy. Notably, the hierarchy derived from our research varies also with Visser's CSR model. © 2023 by the authors. Licensee Bussecon International, Istanbul,


Introduction
The notion that businesses have responsibilities towards society has become more potent than the counter one. However, the nature, extent, and implication of these social responsibilities in different countries seem to be different. Businesses in developing countries are expected to initiate localized CSR practices. African context turns upside-down the CSR pyramid as well. The question to be considered is to what extent the level of development, economic prosperity, or state involvement affects the way businesses perceive and implement CSR.
Ethiopia, a country with its own unique calendar of 13 months, is located in the east of Africa with a population of more than 110 million (ESA, 2017). Agriculture is the backbone of the country. Nevertheless, the service and industry sectors play a key role as a catalyst for the economy. According to the official report of the World Bank, over the last decade, Ethiopia has been one of the fastest-growing economies in the world, with an annual average GDP growth ranging from 7% to 12% (International Monetary Fund, 2016). However, this economic growth came with a higher cost. In the pursuit of attracting foreign direct investments, the government used a loose investment and business policy. On the other hand, big businesses like airlines, transportation, and communication are monopolized by the state. It is expected that the state has additional functions above the conventional role of protecting, regulating, and rewarding. As the state plays a key role in business, the interaction between business and society is worth studying in-depth.

38
In undemocratic state dynamics, governments are unlikely to represent the people's interest (Steiner and Steiner, 2009). It gets worse when the government has a business interest. The policies and actions will tend to favor the government and its business allies. The absence of robust independent media, civil institutions, intellectual freedom, and democratic rights is a suitable environment for exploitive systems to grow and get mature.
In a country where more than 70% of the population is engaged in agriculture for a living, land value is priceless. Unfortunately, in the Ethiopia, the land is owned by the state. Hence, as the state has a stake in the business, it can dismantle the real owners of the land from their settlement without enough compensation either for its own use or for leasing with higher rent at any time (Gebresillasie, 2006).
It is crystal clear that firms do not prosper in a vacuum. Without going into the details about the externalities caused by firms, it is undeniable that many firms in Ethiopia use key resources of the society like land, labor, and other natural resources in exchange for a deficient payment (Worku, 2006). Thus, in addition to the environmental impacts, firms significantly affect society's social and economic livelihood.
Given this assumption, one can ask, did the state's involvement in the business impact the way the businesses behave? More appropriately, did the nature of the social responsibility of businesses differ in this context? Did businesses assume corporate social responsibility practice? If so, at what level is the CSR practice?
Though the millennium development goals crafted in 2000 forced many underdeveloped and developing countries to embed sustainable investment in their pursuit of sustainable growth, CSR did not receive equal attention. This paper will examine the nature of corporate social responsibility practices of homegrown Ethiopian businesses, both private and state-owned. Multinational companies operating in Ethiopia or multinational companies owned by Ethiopians are not included in this study, as their activity is affected by various external factors (Reeve & Pincin, 2018).
In order to achieve the objective mentioned above, the following research questions are outlined: i.
Do Ethiopian companies practice CSR? ii.
What is the nature of CSR practices in Ethiopia?
iii. At which stage are the CSR practices in Ethiopia?
The main contribution of this study is generating a foundation to understand the CSR approaches and practices in a developing country through the example of Ethiopia. While Visser (2006) introduced an African CSR hierarchy contrary to Carroll's (1979) pyramid, this research questions the existence of the same priorities all over Africa. Previous studies reported a newly-emerging positive attitude towards CSR and an infant stage practice. Our work inquiries into the development trend of CSR in Ethiopia. It is also beyond the current studies with its scope due to the coverage of businesses operating in different industries. Moreover, the results will assess the current practices and help Ethiopian businesses improve their CSR practices in different dimensions.
As the nature of this study is exploratory, this study adopts a mix of both qualitative and quantitative research approaches. The qualitative approach involves collecting secondary data from company's websites; organize the data in a manner suitable for the intended analysis. Web-based content analysis is used as part of the qualitative data analysis approach. The quantitative method is employed in both data collection and analysis. A structured questionnaire is distributed, collected, and organized using different survey techniques.
This study consists of five main chapters. The next chapter reviews the literature related to CSR, including its historical development, the stages of CSR, and practices in the African context. Next, the methodology session introduces both exploratory factor and content analysis, which complement each other within our study's context. Finally, while the fourth chapter delivers findings and results, the final chapter concludes the study with a comprehensive discussion.

Conceptual and Theoretical Background The Role of Business in a Society
Though the discussion regarding the role of business dates back to the early years of the 19 th century, it was in the 1930s that the topic got reasonable consideration in the academic circle (Fifka, 2009). Until 1953, the discussion on the role of business towards society was themed by different terms and notions like social engagement and involvement. However, in the year 1953, a scholar named Howard Bowen (1953) introduced the phrase social responsibility in his work "Social Responsibility of Businessmen." In his article, Bowen affirmed that the social responsibility of a business "refers to the obligations of businessmen to pursue those lines of action which are desirable in terms of the objectives and values of our society." Although the idea of the objectives and values of a society is vague and open to different and contradicting definitions and understandings, Bowen asserted that businesses have an irrefutable responsibility towards the society in which they operate.
The famous management guru Peter Drucker (1964) analyzed the mission of management beyond looking after shareholders' interest, underlining their responsibility towards the people, referring to employees, to the country's general economy, and most importantly, to society. Another prominent scholar, Brown (1979), condemned the single-purpose approach as a mere quest for profit-making and called for a change towards a multiplicity of purposes incorporating economic, social, educational, and environmental goals.
On the other hand, few scholars deny the existence of any fundamental social responsibility a business is bestowed on. Their argument revolves around the notion that the very philosophy of business is to make a profit and is inherently an economic entity with a single purpose: maximizing profit (Gallagher, 2010). For scholars like Milton Friedman, making a profit is a sufficient responsibility for any business (Friedman, 1970).
To sum up, while scholars predominantly support the concept of CSR in general, its scope and bottom line are still debatable issues.

Understanding CSR
Though labeled as vague by many scholars (Fifka, 2009), the first definition of CSR belongs to Howard Bowen. According to Bowen (1953), corporate social responsibility refers to "the obligations of businessmen to pursue those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society." However, Joseph McGuire (1963) extended CSR beyond the framework of economic and legal obligations. According to McGuire, "the idea of social responsibilities supposes that the corporation has not only economic and legal obligations but also certain responsibilities to society which extend beyond these obligations" (McGuire, 1963).
Carroll designed a four-level pyramid model of CSR in 1979, which is arguably the most precise and quantifiable conception of CSR. That is why it later became the most established and cited model of CSR. Initially, Carroll defined CSR as "ethical, economic, legal and discretionary expectations that society has of organizations at a given point in time."However, by altering the concept of discretionary expectation into philanthropic acts, he later redefined CSR as a firm's attempt "to make a profit, obey the law, be ethical and be a good citizen" (Carroll, 1979).
Scholars also developed various scales to measure the stages of CSR practice. Johnson (2003) provided a model that helps us to understand the level of CSR practices in a given company. According to him, the first level of CSR is called illegal or irresponsible, where companies do not comply with society's norms and standards and the state's legal requirements. The second level is compliance, where the CSR practice is oriented only toward meeting the minimum legal standards. The third level is fragmented, whereby companies try to engage in activities like charity, community support, and environmental protection beyond complying with rules and regulations. The fourth level is called strategic referring to practices enhancing financial performance. Ethical values embedded in company policies, community support based on need assessment, and environment-friendly business are characteristics of companies at this level. Finally, the fifth level is social advocacy, characterized by CSR practices motivated by intrinsic morals, not profit maximization.

CSR in African Context
Despite the global attention CSR received in the last decades, it has a long journey to go in developing countries (Arli and Lasmono, 2010). Just as business strategies and practices are influenced by culture and context, arguably, CSR cannot escape from this. As Porter and Kramer (2006) pointed out, CSR emerged from external pressure. Given the differences in the market structure, the government, business, and social interaction model, customer awareness, media, and civic institutions' strengths, the perception and attitude toward CSR differ (Matton and Moon, 2008).
CSR was initially understood as some kind of philanthropy work (Porter and Kramer, 2006). This misunderstanding is a prevailing situation in sub-Saharan Africa. Visser (2006) argued that the essence of CSR is quite different in developed and developing countries. Hence, he developed a CSR model for developing countries by extending well-known Carroll's CSR pyramid. According to Visser (2006), the dimensions of CSR proposed by Carroll should be contextualized to Africa's environment. He proposes a CSR model where the economic role of the company is given high priority and importance in the CSR hierarchy due to the high unemployment rate, poor economic conditions, and low investment. In addition, philanthropic activity is second to the economic role in terms of importance due to the prevailing severe economic condition. The importance of legal and ethical responsibilities comes after.

Practices of CSR in Ethiopia
Potluri and Temesgen (2008) studied the attitudes of Ethiopian corporations toward CSR. Their findings showed that there exists a positive attitude towards CSR. However, this attitude does not manifest itself in CSR practices. According to the study, only 40% of the companies exhibit a positive attitude toward information disclosure. Furthermore, while a significant 75% of the public is unpleased with the CSR practices in Ethiopia, 69 % of the employees are unsatisfied with the CSR practices of their companies. In addition, 69% of customers believe that companies manipulate them. Robertson (2009) compared the nature of CSR practices in Singapore, Turkey, and Ethiopia. She used a four-factor scale to determine the developmental CSR stage of each country. According to the study, CSR practices in Ethiopia are at the infant stage, even though the private sector plays a leading role in advancing CSR practices.
Rama, Rao, and Hailu (2016) studied the CSR practices of brewery firms in Ethiopia. The study focused on employees' perceptions of their companies' environmental CSR initiatives. The results revealed that the perception of employees towards the environmental CSR initiative of their respective firms is positive. Besides, the study affirmed that firms gave little attention to disclosing information about environmental management in general and their initiative in particular to external stakeholders regularly.
Deyassa (2016) investigated CSR from an Ethiopian perspective and found out that CSR is a new trend in Ethiopia, which is a result of the increasing inflow of multinational companies to the country and the pressure from NGOs due to the environmental impacts caused by companies. By the way, CSR practices in Ethiopia are mainly philanthropy-oriented initiatives.
Bimir's (2016) study on CSR in the Ethiopian leather industry revealed that firms are engaged in compliance with social, environmental, and economic responsibilities. This compliance is coupled with the state and foreign market pressure as key motivators and the absence of organized CSR plans, which shows a learning process at the emergence stage.
The findings of Elifneh's (2017) qualitative study based on interviews, the companies' documents, and public information indicated that brewery firms in Ethiopia implement common and distinct types of CSR activities towards various stakeholders such as shareholders/owners, employees, local community, customers, farmers, and the environment. Education, health care, employment opportunities for locals, employees' development and learning, improving workplace health and safety, medical and insurance coverage to employees, and women empowerment are among the concentration areas of CSR. The brewery firms also undertake noteworthy environmental practices. However, it might seem that more importance has been given to the social dimension of CSR than the environmental one.
Alemayehu (2017) studied on CSR approach of a soft drink company. It seems the company operates in compliance with local rules and regulations. Moreover, it supports some basic environmental protection activities. However, the lack of an organized and planned CSR initiative makes the company reactive, not proactive, in practicing CSR.
Takele (2018) conducted a study among 152 employees and managers employed by Ethiopian companies operating in the leather industry. The study revealed that companies are mainly focused on CSR to employees and CSR to customers. Nevertheless, salary levels and lack of rigorous control to ensure work safety are among the major concerns of employees. On the other hand, local communities argue for the negligence of companies in environmental sensitivity. For managers, lack of government policy promoting CSR and lack of organizational capabilities are among the significant challenges.
Another study on CSR activities of a multinational brewery company in Ethiopia pointed out how socially responsive it is instead of being socially responsible. Even though it seems the company has successfully integrated CSR into its corporate-level strategy, it has not made a significant effort in order to minimize the environmental damage created by operations in suburban areas. Moreover, the company initiated philanthropic activities after mounting pressure from the residents (Gereziher, 2019).
As introduced above, the number of studies on CSR practices in South Africa, specifically Ethiopia, is limited. Moreover, most existing studies focus on a single industry or region. Another significant limitation is the difference between the CSR approaches of local businesses and MNEs. The general argument of these studies is that Ethiopian business CSR practices must be developed from philanthropy towards sustainable development.

Research and Methodology
This part of the study addresses the different methodologies used throughout this study. It includes research design, sample selection techniques, data collection methods, and the techniques used in data interpretation and analysis.
This mixed-method study includes quantitative analysis through SPSS and qualitative content analysis. In addition, a revised CSR practices scale developed by Turker (2009) has been conducted on owners/managers and employees of 100 local companies in Ethiopia. The original questionnaire is a 5-Likert scale, including 42 statements. Besides, the companies' official websites have been thoroughly reviewed using five variables to understand their CSR approaches and practices better.
The sampling method used in this study is purposive sampling. As there is no secondary stock market in the country, the chamber of commerce and trade associations are the primary source of company information besides the company website. Hence, all the 100 companies selected are listed in the Addis Chamber of commerce and are among the large ones in their respective sector. In order to make the study representative, the 100 companies are selected from various business sectors such as wholesale and retail, technology, agriculture, transportation, financial institution, service, and manufacturing sectors. Multinational companies are not included in this study because multinational companies' policies and actions are affected and driven by international business environment and global trends rather than the domestic business context. As it is hard to receive a sufficient amount of response from mail surveys, collecting data through distributing printed surveys and direct phone calls is preferred.
In order to check the reliability and internal consistency of measurements, the Cronbach alpha test is used. It is a widely accepted rule that results closer to 1 imply greater internal consistency, while results closer to 0 imply little internal consistency. The statistical result shows that the overall Cronbach's alpha is 84%, which is closer to 1.
In this research, descriptive statistical techniques such as mean and standard deviation frequency distribution were used to interpret the data. Data analysis was done using statistical product and service solutions (IBM SPSS, Version 22) and Microsoft Excel. The qualitative data collected from the sample companies' websites are analyzed using content analysis techniques by developing a measurement scale to evaluate the CSR practices reported on each company's websites.

Exploratory Factor Analysis
In this study, Exploratory Factor Analysis is employed by using SPSS. EFA is conducted based on principle component analysis and Varimax rotation (eigenvalue of 1, 25 maximum iterations, suppressed small coefficients below 0.3). Initially, 28 variables were selected to measure the construct. However, four variables were deleted in the process of extracting components. According to the statistical result, the overall Cronbach's alpha is 84%, which is profoundly close to 1 and refers to an acceptable value for reliability. The analysis resulted in six components, as seen in the table.
The variables (questions) in the first component measure CSR activities related to environmental initiatives and philanthropy. In the third component, the loaded variables measure how much a company supports social sustainability. These categories can be unified under the notion of environmental, social, and governmental CSR so it can measure a single construct. The variables in the second component measure management of CSR practices. It is about the vision crafted, the strategies implemented, and the resources utilized to get the maximum output from CSR practices. The variables in the fourth component measure the CSR activity of a company toward its customers. It includes a company's readiness to respect consumer rights and the ethical and legal aspects of providing customers with accurate and complete information about the product and service. The variables in the fifth component measure the CSR activity of a company toward employees. It is measured through the salary and related benefits provided to the employees and the safety of the working environment. The last but not least component is loaded with variables that measure the legal dimension of CSR.

Content Analysis
The content analysis was carried out based on the companies' websites. Five variables were determined to measure CSR practices through annual reports and posts available. Each variable was given a different score with regard to its importance. For example, Visser's CSR model for developing countries prioritizes activities and scores them. The first variable is philanthropy, which covers reports related to the company's donations and sponsorships. The second variable employed is environmental management, which includes environmental policy, impact, and conservation reports. The third variable is the code of conduct, which evaluates how the company interacts with customers, advertises its products, and discloses information and details of the products. The fourth variable is employee safety; it encompasses any ratio regarding the company's working environment. The last variable is sustainable initiatives that measure a company's contribution to sustainable development, like building schools, hospitals, and roads.

Findings
The following section presents descriptive analysis and later data interpretation based on the research questions mentioned above.

Descriptive Statistics
Among the 84 respondent companies, 8.3% of them are from the agricultural sector, 11.9% from banking, 14.2% from construction and real estate, 20.2% from multi-sectoral holding companies, 35.7% from manufacturing, 4.7% from the retail sector, and 4.7% from the service sector. The following figure visualizes the distribution of companies among different sectors. On the other hand, 48% of the companies are public limited companies, while 25% of them are private-limited, and 27% are stateowned and run by the government.

Do Ethiopian Companies Practice CSR?
The first research question of this study is about the existence of CSR practices in Ethiopian companies. The scale introduced by Turker (2006) was carefully re-designed to measure this research question. According to the survey result, only 30% of the respondents have a mean below 2.5. It implies that the vast majority of companies practice CSR. Around 58% of these companies have a mean score between 2.5 and 3, while the remaining 12% have a mean score between three and four. However, none was scored above four. It is a nutshell review of the survey. Hence, it needs a breakdown for a better understanding.

CSR towards Employees
With respect to CSR towards employees, which is measured by the salary and other additional benefits provided by the company, only 2% of the companies have a mean score of 2. Implying that these companies do not practice CSR toward their employees. 7% of the companies have a mean score between 2.5 and 3. Although these companies practice CSR toward employees, it can be assessed as mediocre. On the other hand, 78% of the companies have a mean score between 3 and 4, which is regarded as good.
Last but not least, 5% of the companies have a mean score above 4, which is very good. This result portrays the enormous attention given to employees by the respondent companies. However, the survey results portray another figure. 46% of the employees assessed their company's CSR toward employees as below average. Slightly above this number, 47% of the employee rates their company's CSR toward employees as very good. It seems that what is perceived as reasonable compensation for managers might not be as attractive or good as for employees.

Environmental Social Governmental CSR
The ESG dimension of CSR practice is measured as a combination of environmental management, philanthropy, and social sustainability. The survey result of ESG depicts an acceptable level of CSR practice in this dimension. The total mean result is 2.91. When this figure is broken-down, 36% of the companies have a mean score below 2.5. It implies that these companies almost do not practice ESG-oriented CSR. In the environmental management section of ESG, the overall mean of the companies is 3.04, which is good. In this specific section, 38% of the companies' mean score is below the average of 2.5, implying their poor environmental management. 18% of the respondent companies' mean score is above 2.5 to 3, while 44% of the respondents have a mean score of above 3.
There exists another dissimilarity between managers' and employees' responses regarding ESG-oriented CSR. The average mean score derived from employees is 2.8, which is below the score derived from the managers. This contradiction confirms the skepticism toward CSR and the claim that most of CSR activities are window dressing.
The social sustainability section of ESG is measured by the company's initiatives that promote social well-being and development. In this section, 85% of the companies have a mean score below average, depicting the insufficient attention given to the social wellbeing and societal development aspect of CSR. In addition, while 3% of the companies have outstanding practices promoting societal development, 12% have a good grade. Specifically, 78% of the companies do not have a policy that supports women's recruitment.
As a worth-mentioning finding in this part, 60% of the companies actively participate in donation initiatives. However, without the availability of a well-structured CSR management, this reflects companies' lack of strategy in associating their donations with social well-being.

Customer Dimension of CSR
The respondent companies seem to have a healthier result in the customer dimension of CSR. All companies have a mean score of above 3 to 4, which depicts good practice and attention given to customers. However, this may not give an accurate picture, as this study did not include customers.

Legal Aspect
The other CSR dimension with notably excellent practice is the legal aspect. An outstanding 83% of respondents have a mean score above 3 to 4, and the remaining respondents scored above 4. This result reflects that Ethiopian companies give great attention to the law.

CSR Management
The CSR management component, measured by the CSR department's availability, the existence of a CSR commitment statement in the company's mission statement, attention given to global trends like sustainable development, has a very low mean score. 85% of the respondents have a mean score below 2.5, which is literally bad. Though most of the companies are engaging in selected CSR activities, they failed to have a proper management framework for it. Without proper management, it could probably be difficult to measure the impact of their CSR practice. When the results are summed up, most of the respondent companies have outstanding practices in legal aspects and CSR towards employees and customers. In contrast, their practices in managing CSR initiatives are remarkably poor. Table 2 outlines the assessment of Ethiopian companies' CSR practices based on the dimensions described by this study. Besides, Figure 3 presents a visual assessment of dimensions and average scores. It is proven that most Ethiopian companies practice CSR. However, this is a crude argument that must be filtered. Therefore, the second research question will address this: According to the survey results, the components with the highest mean scores are legal aspect, CSR toward customers, and CSR toward employees, respectively.
We can infer from this result that the nature of CSR practice in Ethiopia is inclined toward employee and customer satisfaction, adhering to laws and regulations, and making monetary donations. CSR practices concerning environmental protection and social well-being are at the infant stage. CSR management is abysmal. Hence, the nature of CSR practice in Ethiopia is quite different from not only the contemporary global trend but also from what Carroll's pyramid proposed. Contrarily, this result correlates with the argument of Visser (2006) that the nature and form of CSR practices in developing countries are different from the well-established developed countries' trends, as the priorities are remarkably different. Interestingly, the hierarchy derived from our study varies also with Visser's model. CSR in the legal aspect takes the lead and philanthropy comes after CSR towards customers and employees.

At Which Stage is the CSR Practice in Ethiopia?
As discussed earlier in the literature review section, different models are used to understand the stage of CSR practices. Johnson's (2003) model of CSR stages is employed for this analysis. According to him, there are five stages of CSR. The first stage is illegal or irresponsible. It is a stage where companies do not comply with norms and legal requirements. The survey result depicts that there is no company at this stage among the respondents. The second stage is compliance, where companies merely strive to meet the minimum legal standards like minimum wage level, environment laws, product safety, and employee health. Results show that all companies take the legal aspect of CSR into consideration and initiate customer-oriented CSR. More than 80% of them clearly prioritize CSR toward their employees. Based on these statistics, it can be argued that Ethiopian companies have largely arrived at that stage. The third stage is fragmented, whereby companies try to engage in philanthropic activities, environmental protection, and so on beyond mere complying with rules and regulations. However, their activity is not well managed and strategic, rather it is fragmented. 60% of the respondent companies contribute to charity activities. More than half of the companies take an environmentalfriendly position. By the way, the rate of companies executing a well-structured CSR Management is just 6%. Despite the fact that many Ethiopian companies have obtained the third stage, the stage of strategic CSR seems distant for many of them. In this stage companies' CSR is well managed in a way to enhance the financial performance of the company. The fifth stage is social advocacy, which is an organizational philosophy whereby profit-making is not the sole objective of the company. At this stage the driver of CSR practice is not financial return, rather it is an intrinsic value of striving for social well-being. None of the respondent companies is at this stage. As more than half of the companies are at the fragmented stage, it can be generalized that the practice of Ethiopian companies is at the stage of fragmented CSR.

Results from Content Analysis
The content analysis is carried out based on companies' websites. Five variables are used to discover CSR practices from reports and posts available on the respective website. Visser's CSR model for developing countries is used to prioritize activities and score them. Among the 84 respondent companies, only 58 have a website. Hence, the interpretation of this content analysis is based on 58 companies, which is 70% of the total company sample.
The analysis of the websites firstly revealed that, even if there is no standardized CSR report disclosure in most companies, activity reports make CSR approaches and practices of companies visible.
Results show that philanthropy is a common practice among companies. 94% of the companies have a specific philanthropic activity. The second dimension is environmental management. 46% of the companies have published environmental management reports on their website. It indicates either these companies' failure to report their environmental practices or their poor environmental practice.
The third dimension is related to the code of conduct that discovers companies' legal relations with customers, the government, and competitors. It seems that Ethiopian companies give due attention to that. The fourth dimension is the employee safety report. 77% of the companies disclose reports related to the working environment safety. The last dimension is the sustainable activity report. The result indicates that only 18% of the companies disclose reports related to their sustainable development activity.
The total score of Ethiopian companies is 68 out of 100. However, findings also indicate that some companies undertake certain activities that are regarded as CSR even if they do not use the term CSR.
The following Figure 4 compares our quantitative study's results with the content analysis findings. It clearly shows that both methods introduce very similar rankings for Ethiopian companies' CSR practices, except for philanthropy.

Discussion
This study aimed to examine the CSR approaches and practices of Ethiopian firms. The study's finding corresponds to some similar studies of the subject matter in Ethiopia. Potluri and Temesgen (2008) studied the attitude of Ethiopian firms, and their finding depicts an overall positive attitude toward CSR. Another study by Rama and his fellows (2016) reflects similar results about CSR practice in Ethiopia.
According to the findings from the survey and content analysis, it can be concluded that Ethiopian companies operate at the stage of fragmented CSR. On the other hand, Bimir's (2016) study on the leather industry discovered CSR practice at the compliance level.
Many recent studies (Alemayehu, 2017;Elifneh, 2017;Gereziher, 2019;Takele, 2018) indicated negligence in the environmental sustainability practices of Ethiopian businesses. This study revealed a similar result that could be clearly observed through the CSR pyramid in Figure 4.
In general, most of the respondent companies have outstanding practices in the legal aspect, CSR towards employees and customers, and philanthropy. However, social sustainability and management of CSR are remarkably mediocre. According to the study, the CSR practice in Ethiopia is not well managed and value creation oriented. What is remarkable is that all previous studies conducted in the same region point out a very similar result: lack of an organized and planned CSR initiative prevents businesses from practicing CSR proactively.
Nevertheless, the CSR practice in Ethiopia is quite different from what is argued in theory and practiced within the framework of the global trend. This difference is both expected and acceptable because of a different structure of stakeholders, pushing actors, corporate governance, economic status, and culture.
All companies overwhelmingly respect rules and regulations as a CSR dimension in the broad spectrum. CSR towards employees has also been given considerable attention. Hence, it can be inferred that the orientation of CSR practice in Ethiopia is directed toward respecting rules, regulations, and employee rights. It is pretty different from the CSR trend of other developing countries and the model developed by Visser (2006). A similar study in neighboring Kenya showed that Kenyan companies prioritize philanthropic activities (Muthuri and Gilbert, 2011). A study in South Sudan also portrayed that CSR is oriented toward philanthropy (Ives and Buchner, 2011). Another study in Angola reflected a similar trend of prioritizing philanthropy (Luiz Cortes et al., 2014). Ehir (2016) questioned the relevance of Visser's model in the example of Nigeria and empirically validated Visser's hierarchy that starts with economic responsibilities followed by philanthropic, legal, and ethical responsibilities.

Conclusion
From a comparative point of view, the nature of CSR practices in Ethiopia is quite different from the contemporary global trend and what Carroll's pyramid proposed. Contrarily, this result correlates with the argument of Visser (2006) that the nature and form of CSR practices in developing countries are different from the well-established developed countries' trend, as the priorities are remarkably different. Interestingly, the hierarchy derived from our study varies also with Visser's model. CSR in the legal aspect takes the lead, and philanthropy comes after CSR towards customers and employees. A content analysis conducted to enhance our understanding of Ethiopian CSR is similar to the results of the quantitative study, excluding philanthropy. By the way, it can be generalized that the CSR practice of Ethiopian companies is at the stage of fragmented CSR according to Johnson's model.
Even though CSR practice towards customers is relatively better, it needs improvements beyond merely listening to customer complaints and respecting their rights to enhance customer loyalty. Higher customer loyalty increases profitability and creates an opportunity for sustainable growth. In addition, this may also improve brand image.
Another dimension that has to be improved is CSR toward employees. In this dynamic business world, having a competent and proactive employee is the key to achieving organizational goals and overcoming ever-increasing competition. Hence, companies should tailor their CSR toward employees to improve their living standards and enhance their capacity.
As noted before, CSR management is mediocre in Ethiopia, reflecting little attention given to proper CSR management. Hence, it is recommendable to have a well-organized management team and a solid infrastructure to bring the best out of CSR practices. It may also help to utilize resources that are allocated to CSR properly. Another underdeveloped dimension of CSR in Ethiopia is poor reporting and disclosure. Though there are web-based reports including ethical, legal, and philanthropic issues, it is highly recommended to disclose all CSR practices timely, using a proper channel, and in a well-organized and standardized format. It is not only creates public trust but also enhances accountability and transparency. Based on the recommendation of scholars like Visser (2006), the philanthropic aspect of CSR is of high value in the developing world. Hence, special attention should be given to strategic and coordinated philanthropy based on need assessment and sustainable development.
The environmental aspect of CSR is also another dimension given little attention. Companies need to work in collaboration not only to safeguard the environment but also to improve the overall ecosystem. The government should enforce strict rules and regulations to create environment-friendly business dynamics. Media, NGOs, and other local and global advocacy groups should also lead in creating awareness and lobbying for environmentally-friendly and sustainable investments. It will contribute not only to a safer and livable environment but also to the prosperity of the country.
It should be mentioned among the limitations that the companies included in this study are only large companies and those that are members of the chamber of commerce. Measuring only the CSR practice of large companies may not precisely reflect the overall CSR practices of Ethiopian companies. Hence, including medium and small-scale companies and studying their perception of CSR could be a future direction of study.
This study could be enlarged in different directions. First, the study sample could be enriched by firms from different industries or regions in Ethiopia. Studying the differences between the practices of various business sectors is also a related area for further study. An increase in sample size certainly increases the quality of the findings.
Instead of conducting research based on employee perceptions, CSR activities could also be assessed from a customer perspective. On the other hand, measuring the impact of CSR practice in Ethiopia is also a subject area given little attention.
Finally, the theoretical framework behind the study includes Carroll's and Visser's pyramids. However, Africa has its unique historical background of colonization and cultural codes, such as Ubuntu. Therefore, researchers should analyze CSR in Africa from a culturespecific perspective and offer development suggestions accordingly.
To sum up, the main contribution of this study is generating a foundation to understand the CSR approaches and practices in a developing country through the example of Ethiopia. While Visser (2006) introduced an African CSR hierarchy contrary to Carroll's (1979) pyramid, this research questions the existence of the same priorities all over Africa. Previous studies conducted over a decade ago reported a newly-emerging positive attitude towards CSR and an infant-stage practice. Our work attempted to inquire into the development trend of CSR. It is also beyond the current studies, with its scope covering businesses operating in different industries. Moreover, the results assess the current practices and help Ethiopian businesses improve their CSR practices in different dimensions.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to restrictions.

Conflicts of Interest:
The authors declare no conflict of interest.