“Integrating ESG factors in investment decisions by mutual fund managers: a case of selected Johannesburg Stock Exchange-listed companies”

This paper examines whether mutual fund managers incorporate environmental, so- cial, and governance (ESG) factors when deciding which sector to invest on behalf of their trustees. In doing this, the top 20 South African mutual fund companies (asset managers) listed on the Johannesburg Stock Exchange (JSE) were selected. The paper identified the top 30 JSE listed companies (in the large industrial, equipment, and ma- chinery sectors, excluding unlisted and service-oriented companies) where trustees’ funds were invested (with a total of 28 companies between 2007 and 2017) from the mutual fund companies’ Equity Fund Fact Sheets 2017 (representing recent investment focus). ESG data were collected from the integrated and sustainability reports at the sampled companies’ websites, and financial data were sourced from the IRESS database. This study adopted the panel data analysis. The results show an insignificant negative relationship between the ESG proxies (water usage, employee health and safety cost [number of work-related fatalities], percentage of women on corporate board) and return on equity (ROE). This means that the sampled companies disregard the United Nations Principle of Responsible Investment (UN PRI) guideline, suggesting that asset managers focus on increasing returns on shareholders’ investment without considering ESG issues. The paper concludes that the disregard for responsible invest- ment guidelines does not encourage companies to improve their unsustainable business practices.


INTRODUCTION
In the face of the current global business climate, non-adherence to responsible investment practices could pose risks of both reputational damage and consumer backlash, thereby exposing businesses to disruptions and spiraling costs. By integrating both financial and ESG factors in investment decisions contribute to sustainable business practices (Stankevičienė & Čepulytė, 2014). Companies seeking to achieve sustainable business practices cannot succeed by focusing on individual aspects of the ESG factors; instead, they need to integrate them to attain corporate strategic objectives (Korditabar, 2015;Escrig-Omedo et al., 2017). Likewise, companies seeking long-term sustainability should consider ways of integrating these non-financial (ESG factors) into their core business decisions. By integrating ESG issues in investment decisions, companies are more conscious of the long-term impact of their operations on the environment and society (Husted & Milton de Sousa-Filho, 2017; Zulkafli et al., 2017). The principle of re-sponsible investment advocates that companies aiming to be sustainable should focus beyond financial factors in making investment decisions. Responsible investment entails the inclusion of non-financial factors (ESG) into the choice of investment opportunities (Stankevičienė & Čepulytė, 2014). In being consistent with the triple bottom line (TBL) theory, the inclusion of ESG factors will promote sustainable business practices (Elkington, 1994;Korditabar, 2015). Likewise, Pilaj (2017) reiterates that responsible investments explicitly contribute to companies' sustainable business practices. Disregarding the United Nations Principle of Responsible Investment (UN PRI) guideline suggests that asset managers focus on increasing returns on shareholders' investment without considering ESG issues. However, most mutual fund managers do not consider environmental, social, and governance (ESG) issues in making investment decisions. Regarding environmental issues, Korditabar (2015) contends that companies should be committed to using water and other natural resources efficiently and sustainably because unsustainable use of water might affect future generations' ability to meet their water needs. Likewise, companies adhering to sustainable business practices will attract and keep good employees (Auer, 2016). When a company invests responsibly in employees' health and safety, such employees are likely to feel safe and secure at work. On the other hand, employees may feel unsafe when management focuses less on improving sustainable business practices. Therefore, it is vital to empirically examine the impact of conscious investments (Rao & Tilt, 2016) in employees' health and safety costs and sustainable practices in board composition, such as the percentage of women on corporate boards on the financial performance of companies. Therefore, it is crucial to examine whether the profit motive drives mutual fund managers' investment decisions over and above social and environmental considerations. This study also measures companies' financial performance of investee companies using the return on equity (ROE).
Hence, this study examines whether mutual fund managers integrate ESG issues when making investment decisions on behalf of their trustees.
The study consists of the following sections: literature review, description of the study methodology, results and discussion, and conclusion. . Firm size may be among the factors that influence the company's financial performance and is, therefore, included in this study as a control variable. In terms of the relationship between firm size and financial performance, earlier studies have shown inconclusive evidence. A study conducted in the Nigerian manufacturing sector found a positive relationship between firm size and profitability using both total assets and total sales as measures (John & Adebayo, 2013). Likewise, a study conducted in the Croatian manufacturing sector also found a positive relationship between firm size and profitability (Pervan & Višić, 2012). A study performed in Pakistan using companies listed on the Karachi Stock Exchange also found both weak and positive relationship between firm size and financial performance (Abbasi & Malik, 2015). However, a study performed on listed manufacturing firms in Sri Lanka found no relationship between firm size and profitability (Niresh & Thirunavukkarasu, 2014). Thus, this paper controlled for the effect of firm size on financial performance. In terms of other ROE determinants, factors identified include tax burden, interest cover, operating margin, asset turnover, financial leverage, price to earnings, the book to market, and current ratio (Kharatyan et al., 2017) as control variables. Determinants such as tax burden, interest burden, operating margin, asset turnover, and financial leverage ratios have a significant effect on ROE (Kharatyan et al., 2017). In the same study (Kharatyan et al., 2017), the effect of price to earnings, book to market, and current ratio on ROE were found not to be significant. The study by Kijewska (2016) supported the use of operating profit margin, asset turnover, financial leverage, and tax ratios as significant determinants of ROE.

LITERATURE REVIEW AND HYPOTHESES
The above extant literature shows the inconclusiveness of results from researchers. Besides, fewer studies have attempted to combine the selected proxies. This study examines the relationship between environmental sustainability (water usage), social sustainability (employees' health and safety cost) and governance issues (the percentage of women on corporate boards), and the financial performance of selected JSE listed companies; which mutual fund managers have invested trustees' fund. Thus, this study contributes to the debate on responsible investment and sustainable business practices by combining the effects of ESG factors on South African companies' financial performance where mutual fund managers invest trustees' funds.
Hence, this paper considered the following hypotheses: H 1 : There is no correlation between environmental sustainability investment (water usage) and ROE.
H 2 : There is no correlation between social sustainability investment (employee health and safety cost) and ROE.
H 3 : There is no correlation between the percentage of female board members and ROE.
H 4 : There is no correlation between firm size (market capitalization) and ROE.
H 5 : There is no correlation between ROE influencers and ROE.

METHODOLOGY
The paper analyzed the data collected from 28 companies listed on the JSE where mutual fund managers invest between 2007 and 2017 using panel data analysis. Besides, the sampled companies' operations are considered to affect both environment and society adversely. Although there is a stock of over 350 different companies listed on the JSE, the sample was narrowed to only the selected 28 manufacturing companies where mutual fund managers invest. This is because of their use of a large volume of water in their production processes and the industrial hazards posed to their employees from consistent use of heavy equipment and machinery. Besides, this paper excluded unlisted companies for lack of data availability and service-oriented companies because they are considered to use less volume of water. This study used annual, integrated, and sustainability reports by companies from the top 30 listed JSE companies where South African mutual funds (investment fund managers) invest trustees' funds. This was done to determine whether the mutual fund's companies consider ESG factors in their investment decisions in promoting sustainable business practices among JSE listed companies. The mutual fund companies' equity funds fact sheets for the 2017 financial year-end (because it represents recent investments made) was used to select the mutual fund companies and the top 30 companies in which they invested trustee funds. This paper arrived at a sample of only 28 companies because our data search was narrowed to only those companies where mutual fund managers have consistently invested trustees' funds in the period of investigation. Besides, one excluded those JSE listed companies with incomplete data set identified for this study. The sampled companies cut across various industries that include basic materials, chemicals, consumer goods, general industries, health care, mining, tobacco, pharmaceuticals, and properties. However, the sample excluded companies considered as having a less adverse impact on environmental sustainability such as banks, financial services, insurance companies, media, mobile telecommunications, technology, and because they use a lower volume of water and are low on the use of heavy industrial equipment and machinery. The choice of the limited sample was because the intention was to examine whether mutual fund managers consider ESG issues when making investment decisions on behalf of trustees. This paper sampled only those companies in which mutual fund managers have consistently invested their trustees' funds and so determine if these mutual fund managers consider investing responsibly (that is, using ESG criteria as the motive for investing or are still investing to increase trustees' returns on equity (ROE)). The study extracted data for variables such as water usage as a proxy for environmental performance; employee health and safety cost (investment to reduce employees' work-related hazards or injury) as a proxy for social performance; and the percentage of women's representation on corporate boards as a proxy for governance. Besides, the study used ROE as a proxy for companies' financial performance. The ESG variables were extracted from sampled companies' integrated reports (annual reports and sustainability reports) on their websites. The ROE data was sourced from the IRESS database. Hence, the following study model was presented: while water usage, employee health, and safety cost, and percentage of women's representation on corporate' boards are the independent variables. Control variables used in this study include firm size (market capitalization) and ROE influencers.

RESULTS
The risk of reputational damage to and consumer backlash against companies that fail to pay substantial attention to ESG issues in making investment decisions could expose such companies to business disruptions and spiraling costs. Descriptive statistics are presented in Table 1.
As shown in Table 1, the ROE has a mean of 16.82% and a standard deviation of 33.5, meaning that the data are slightly inconsistent and is affected by outliers as the maximum observation is 126.4%. Water usage has the highest mean of 274.226, showing that most of the companies that mutual funds have invested in use water excessively. In terms of the number of work-related fatalities, the mean is 3.34 and a standard deviation of 7.02, which is consistent. Whereas, the percentage of women on the boards has a mean of 10.22%, and the maximum observation representing the percentage of women on the board of a particular company is 50%, implying that few companies are complying with gender equity requirements in their board structures. The regression analysis result is in Table 2. The authors used the cross-sectional time-series FGLS regression and the generalized least squares coefficient where the variance of the residual term is heteroscedastic with no autocorrelation.
The results in Table 2 show no autocorrelation between the variables. The results show that prob>chi 2 is significant (prob > chi 2 = 0.0000) at less than p-value of 0.05. This shows that the variables of this study are significant in interpreting the results. The results show an insignificant negative (p > z =0.834) relationship between water usage and ROE. This signifies that more water usage results in a decrease in ROE and vice versa. The results further indicate that there is a significant (p > z =0.018) and negative relationship (coefficient = -0.19291) between employee health and safety (number of work-related fatalities) and ROE, implying that more fatalities result in a decrease in ROE and vice versa. Moreover, there is an in- Furthermore, the results show a negative relationship between market capitalization (coefficient = -0.04764), book value per share (coefficient = -0.00026), current ratio (coefficient = -4.42256), and ROE indicated by a coefficient of variation. This shows that when ROE increases, these variables decrease. Lastly, the results shows a positive relationship between interest cover (coefficient = 0.033922), leverage factor (coefficient = 0.754592), operating profit margin (coefficient = 1.384349), price-earnings (coefficient = 0.028005), total asset turnover (coefficient = 7.317292) and ROE. This shows that ROE increases when these variables increase.
Data used should be normally distributed in order to avoid distorting the assumptions that apply to regression analysis. If the data is not normally distributed, it can result in unreliable and invalid research results. Hence, the study used the Shapiro-Wilk W test to test for normality in Table 3. Based on the information shown in Table 3, the study data is normally distributed, as indicated by p-value (0.000). The regression of assumption relating to normality is not violated at all. Therefore, the statistical tests are not distorted; and this confirms the validity and reliability of the results.
In confirming whether multicollinearity exists between the variables, variance inflation factor (VIF) tests were conducted, as shown in Table 4. The mean VIF (1.01) shown in Table 4 indicates that the variables' multicollinearity score is not statistically significant to distort the results as the population coefficient can be precisely predicted. Therefore, the study results are acceptable.
Results were interpreted applying the random-effects model (see Table 5) after a Hausman test (see Table 6) was conducted to determine an appropriate model between fixed-effects and random-effects models. The latter proved to be appropriate. Table 5 shows a significant p-value (prob > chi 2 = 0.000), which is less than the significance p-value of 0.05. Likewise, the results show an insignificant negative relationship between water usage, the number of work-related fatalities, the percentage of women on the corporate board, and ROE. Furthermore, the results show a significant neg-ative relationship between market capitalization and ROE. Moreover, the results show an insignificant negative relationship between book value per share, current ratio, interest cover, price-earnings, and ROE. Lastly, the results show a significant positive relationship between leverage factor, operating profit margin, total assets turnover, and ROE. The discussion is based on the random-effects model analysis, as shown by the Hausman test results.
The study accepts the null hypothesis based on the Hausman test that showed that the random-effects model is appropriate, while the study rejects the alternative hypothesis, the fixed-effects model, in interpreting the result. The significant p-value is at 0.05. If the Hausman test results are less than the significance value, the study rejects the null hypothesis while accepting the alternative hypothesis. However, with a Prob > chi 2 = 0.1908 in Table 6, which is more than the significant p-value of 0.05, the null hypothesis is accepted, and the random-effects model (REM) was applied to interpret the results.

DISCUSSION
The result relied on the random-effects model analysis as the appropriate model based on the Hausman test in Table 6 for the discussion. The results in Table 5 show an insignificant negative relationship between water usage, the number of work-related fatalities, the percentage of women on corporate boards, and ROE. Further discussion on the results was based on the hypotheses.
H 1 : There is no correlation between environmental sustainability investment (water usage) and ROE.
A negative relationship exists between water usage and ROE. This shows that the selected companies use water inefficiently or unsustainably, thereby impacting ROE negatively. A large volume of water consumption is associated with a higher cost, which negatively affects profit and ROE. The result is similar to earlier studies with a positive relationship between environmental sustainability measures and firm performance The results show that mutual fund companies overlook sustainable business practices like efficient water usage in selecting investment opportunities. This implies that mutual fund managers do not sufficiently consider environmental issues when making investment decisions but focus more on the returns accruable to trustees (Manzhynski et al., 2015). Besides, the four-capital model postulates that companies should integrate financial performance measures with social and environmental factors to achieve sustainable business practices (Bojan, 2007 However, the result is consistent with earlier studies whereby companies prefer the profitability objective over meeting the needs of the different stakeholders (Friedman, 1970;Shim, 2014). This indicates that mutual fund managers follow Theoretically, the more the company uses the assets, the more the book value is expected to decrease, and consequently, ROE increases. The negative relationship regarding the current ratio, interest cover, and ROE does not make logical sense. One would expect companies with high-interest cover and current ratio to make high ROE. On the other hand, the results show a significant and positive relationship between leverage factor, operating profit margin, total asset turnover, and ROE. Likewise, the positive relationship between price-earnings and ROE is insignificant. This result supports other earlier studies that a positive relationship between these ROE 'influencers' and ROE (Kijewska, 2016;Kharatyan et al., 2017). This implies that some of the selected companies with high assets turnover and operating profit margin are likely to have a high ROE. Thus, this paper rejects the null hypothesis and accepts the alternative hypothesis.

CONCLUSION
This study examined whether South African mutual fund companies considered selected ESG factors in investment decisions and the effect of the selected factors on financial performance. This study found that in making investment decisions, the selected mutual fund companies did not consider ESG factors. The proxied ESG factors have an insignificant negative relationship with firm performance. This is a sign that careful consideration of these factors can result in improvement in financial performance. Investment fund managers should consider educating the trustees, especially on the African continent, about the importance of incorporating ESG issues in investment decisions and of the benefits that emanate from such investments. If investors are at the forefront of sustainable business practices, it may force the investees to start taking ESG issues seriously by becoming aware of losing investments if they do not. Future studies could consider incorporating ESG indicators other than the proxies used in this study to examine responsible investment patterns among mutual fund managers.