Tax Avoidance and Corporate Social Responsibility Disclosure in Ghana.

Purpose: This study seeks to bring new contributions to the literature, by investigating the relationship between CSR dimensions and corporate tax behaviour in a sub-Saharan African country: the Ghanian context. The social aspects of corporate taxation have attracted notable attention in the academic literature over the past ten years, resulting in a prolific stream of research exploring the link between CSR and tax avoidance, but producing conflicting evidence. While several studies have reported a negative relation between CSR and tax avoidance (e.g., Gulzar et al., 2018; Laguir et al., 2015; Lanis & Richardson, 2012, 2015; López-González et al., 2019; Ortas & Gallego-Alvarez, 2020; Salhi et al., 2019), other studies found evidence that firms claiming to be socially responsible actually avoid taxes (e.g., Alsaadi, 2020; Davis et al., 2016; Hoi et al., 2013; Landry et al., 2013; Lin et al., 2018; Col & Patel, 2019; Watson, 2015; Zeng, 2019). The majority of studies have been conducted in developed contexts, and a recent scientometric analysis suggests that African countries remain underexplored (Issah & Rodrigues, 2021). Early studies in developing countries highlighted the limited existence of formalised CSR initiatives (Jamali & Karam, 2016). Yet, according to Sorour et al. (2021), there has been a significant uptick in levels of CSR engagement (operationally and strategically), that seems to be supported by a more complex, varied and nuanced set of motivations. Thus, CSR in developing countries is an emerging field of study (Jamali & Karam, 2018). Formed from the British colony of Gold Coast and the Trust Territory of Togoland in 1957, Ghana became the first sub-Saharan country to gain its independence and it has been a stable democracy since 1992 (Heritage Foundation, 2022). According the World Bank, in the past two decades, Ghana has taken major strides toward democracy being consistently ranked in the top three countries in Africa for freedom of speech and press freedom.14 Yet, the Ghana Integrity Initiative report that Ghana scored 43 (out of 100 points) and ranked 73 out of 180 countries in the Corruption Perception Index for 2021, but failed to make progress in the fight against corruption.15 Ghana is Africa’s second-biggest producer of gold and second largest producer of cocoa in addition to being rich in diamonds, timber and oil (Heritage Foundation, 2022). The World Bank data reveals a fast economic growth of Ghana (except for the COVID-19 pandemic period), but also highlights a poverty rate of about 25 percent. According the United Nations Conference on Trade and Development (UNCTD, 2019), the fast economic growth indicates the attraction of foreign direct investments (FDI), with multinationals coming into the country to take advantage of the tax and other incentives available in specialised economic zones (SEZ), called Ghana Free Zones (GFZ) (e.g., firms located in GFZ benefit from a 100% exemption from payment of income tax on profits for 10 years which will not exceed 15% thereafter), and a stable political environment. The UNCTD (2019, p.191) noted that “Ghana has one of the more successful SEZ programmes in Africa today”. Methodology: Given the context of Ghana, the lack of studies on the association between CSR and tax avoidance in developing countries, in this study we test the following two research hypotheses: H1: Disclosures on the CSR dimensions (economic, environmental and social) are associated with the extent of corporate tax avoidance behaviour in Ghana. H2: The location of companies in Ghana’s Free Zones mediates the relationship between disclosures on the CSR dimensions (economic, environmental and social) and tax avoidance in the country. This study uses a sample of 252 firm-year observations (70 unique companies) of listed (Ghana Stock Exchange) and unlisted firms operating in Ghana, for the four-year period 2017-2020. To test H1 the research model includes as dependent variable tax avoidance; as explanatory variables the CSR dimensions (economic, environmental and social); as controls a set of variables commonly used in prior tax avoidance literature (firm size; capital intensity; leverage; return on assets; dummy==1 for negative profits in prior year; intangibles; auditor type; ratio of executive directors to total board directors; dummy=1 for listed firms; dummy=1 for local firms; dummy=1 for firms located in Ghana Free Zones), and industry and year fixed effects. To test H2, the research model includes as dependent variable tax avoidance, and as independent variables the same as in the prior model, as well as the interaction of the dummy=1 for firms located in Ghana Free Zones (=0 otherwise) with all the independent variables. Two different measures are used as a proxy for tax avoidance. One is the cash effective tax rate (Cash ETR), which corresponds to the cash taxes paid (as shown in the cash flow statement) divided by pre-tax accounting (book) income. ETR measures denote the proxy measure of tax aggressiveness most frequently used by academic researchers (e.g., Davis et al., 2016; Hoi et al., 2013; Lanis & Richardson, 2012, 2015). To allow for meaningful interpretation, values of Cash ETR are truncated between 0 and 1 (see, Davis et al., 2016; Zeng, 2019 for a similar approach). The second measure is the tax rate difference (Tax-Dif), computed as the difference between the statutory tax rate and (cash) effective tax rate. TaxDif captures the deviation from the fair share of tax payable (Lin et al., 2018). Given that higher Cash ETR represent a proxy of lower corporate tax avoidance, we transformed TaxDif by multiplying it by –1 to obtain a measure where higher values suggest less tax avoidance (e.g., Lanis & Richardson, 2012). The CSR variables (CSR-economic, CSR-environmental, and CSR-social) are computed based in a content analysis approach and a word count method performed on annual financial reports and stand-alone CSR reports of listed and non-listed firms in Ghana, using GRI standards as a structural benchmark (from the 34 sub-topics of the GRI standards, through the content analysis, we only found information for 20 sub-topics). Content analysis is a popular method to undertake methodical, and quantitative analyses of content communication of CSR (Feng & Ngai, 2020). The word count method is used because words provide a more controllable analysis (Gao et al., 2005) and word-frequency measures produce both powerful and replicable results in commonly studied settings for measuring disclosure of qualitative information (Henry & Leone, 2016). To develop measures of the quantity of CSR disclosures in Ghana, an equal-weighting scheme, which signifies raw word counts, has been applied, where all the attributes are given the same weight (Kulshrestha & Patro, 2021 followed a similar approach). The firm-year CSR variables represent the sum of the number of words obtained from the word counting process for each of the 20 GRI subtopics. In particular, CSR-economic is given by the sum of the number of words of the 4 related GRI sub-topics; the CSR-environmental corresponds to the total words of the 6 related GRI sub-topics, and CSR-social is equal to the sum of the number of words of the 10 related GRI sub-topics. Alternative measures are also computed by transforming the firmyear values of the CSR dimensions into rankings and dividing it by the total number of firms of the year, resulting in normalized scores ranging between 0 and 1. Results: OLS with year and industry fixed effects and robust standard errors is used to estimate regressions and test the hypotheses. Preliminary results reveal that the coefficient of CSR-economic is negative but statistically insignificant; the coefficient of CSRenvironmental is negative and statistically significant at the 1% level; the coefficient of CSRsocial is positive and statistically significant at the 1% level, both when Cash ETR and TaxDif are used. Qualitatively similar results are obtained when CSR variables are computed in a ranking basis. The results, thus support H1. The results suggest that Ghanian firms disclosing higher levels of environmental information present higher levels of tax evasion. The evidence collected on the environmental disclosure is a bit surprising given the legislative instruments in Ghana for sensitive industries on their environmental impact and spearheaded by Ghana's Environmental Protection Agency (EPA). EPA requires periodic reportage on entities environmental impact, and entities that do not comply can face punitive sanctions, damaging their reputation. A possible reading is that entities disclose environmental information to avoid sanctions and damage their reputation (Welbeck et al., 2017). The evidence collected is consistent with that of Fallan and Fallan (2019) and Laguir et al. (2015), supporting the idea that the more entities' engagement in voluntary environmental disclosure, the greater the tax aggressiveness. Contrarily, firms reporting higher levels of social information present lower levels of tax avoidance. This finding suggests that as entities engage more in social activities (such as public policy and donations, local community support, employee education, health and safety, customer health and safety), the higher is the effective tax rate they pay. This results is consistent with that of Laguir et al. (2015) and Kiesewetter and Manthey (2017). The positive association between the social score and Cash ETR (Tax-Dif) fits the common understanding of social responsibility, and the interpretation implies that social incentives could help to tackle tax avoidance (Kiesewetter & Manthey, 2017). Regarding the test of H2, results for firms located outside the Ghana’s Free Zones are qualitatively similar that reported in the prior paragraphs. Yet, firms located in Ghana’s Free Zones present a statistically significant positive relation between CSR-environmental and Cash ETR (Tax-Dif), suggesting a “green” social responsibility and maybe that the legislative instruments for sensitive industries may be more effective in Ghana’s Free Zones. Furthermore, we found a statistically significant negative relation between CSR-social and Cash ETR (Tax-Dif), suggesting that firms located in Ghana’s Free Zones seems engage in voluntary social disclosure to avoid sanctions and damage their reputation. Ghana Free Zones seem to have a mediating role in the relationship between CSR disclosure and tax avoidance. The results, thus support H2. Research limitations: The inherent subjectivity of the content analysis approach and the word count method used to compute the CSR variables is seen as a limitation of this study. Originality: Over the past ten years, a prolific stream of research exploring the link between CSR and tax avoidance has been undertaken, particularly in economically developed contexts, but the evidence gathered has yielded conflicting results. This study expands the current literature’s focus on developed economies by examining the association between CSR and tax avoidance in the setting of a developing African economy, i.e., Ghana, and by exploring how particular features of the Ghanian context may mediate those relationship. The results may shed some light about the need of looking at the CSR reporting in developing countries as a possibility of a blend of instrumental and normative motivations as suggested by Sorour et al. (2021). In particular, this study explores the mediating effect of Ghana’s Free Zones on the association between the CSR and tax avoidance. [ABSTRACT FROM AUTHOR]