Voluntary disclosure of listed banks operating in Ghana

Issues of voluntary disclosure have become relevant in Ghana after the Bank of Ghana took steps to rationalised banks in Ghana to strengthen the banking sector. The sudden collapse of some banks in Ghana may have raised some doubts among investors and other stakeholders with regards to the quality of information disclosed by banks in Ghana. It has therefore become important for banks in Ghana to improve the quality of financial information by setting up good governance structures and to provide adequate voluntary information in their annual reports to boost the confidence of investors and other stakeholders. This study examines the extent to which listed banks in Ghana voluntarily disclose information in their annual report for the 2018 period and to determine the differences in the types of information voluntarily disclosed by listed banks in Ghana. This study revealed an overall voluntary disclosure level mean of 0.6294 which was slightly above average. The Tukey post hoc test showed a statistical significant difference between strategic information and non-financial information as well as a statistical significant difference between strategic information and financial information. However, there is no statistical significant difference between non-financial information and financial information. The study concludes that adequate voluntary information must be provided in the annual reports of listed banks in Ghana to ensure transparency, consistency, reliability and comparability to influence the economic decision needs of investors and other users of accounting information.


Introduction
Companies desire to raise capital from investors in the capital market necessitates the provision of voluntary accounting information (Choi, 1973). This motivate companies to willingly provide excess non statutory information in their annual report to be useful to several Issues of disclosure have been relevant to a developing country like Ghana on the aftermath of the rationalisation of the banking industry by the central bank which ensued in 2017 and 2018 to sanitise and strengthen the banking sector. The rationalisation of the banking sector resulted in some banks been merged whereas other banks had their licences revoked and this may have raised doubts among investors as well as other stakeholders with regards to quality of details disclosed by these banks. The recent collapse of several operating banks in Ghana has triggered the need to assess the level of voluntary details disclosure by these banks. However, previous studies have not made a thorough assessment on the extent to which listed banks operating in Ghana voluntarily disclose information in their annual report on the aftermath of the central bank's rationalisation of the banking industry. For this reason, there is a massive information gap and this study fills that gap by assessing the level of voluntary details disclosure by listed banks operating in Ghana and proceeds to determine the variations in the types of voluntary details disclosure by listed banks operating in Ghana.
This study reviewed available literature to examine empirical research on disclosures of companies and what other researchers say with regards to recent level of disclosures in firm's annual reports.
The purpose of this study is to provide empirical evidence on the extent to which listed operating banks in Ghana voluntarily disclose details in their annual reports and the variations in the types of voluntary details disclosed to assist potential shareholders and other accounting users in making informed economic decisions.
supplementary information subject to audit and other regulatory scrutiny, intended to be used by management but considered to be pertinent to the economic decision exigency of users of accounting details. Providing adequate voluntary details assist users to have a fair view of what goes on in a company and also benefit companies and the general economy of a country as well as lowering companies cost of capital (Leuz & Verrecchia, 2000;Choi & Levich, 1990;Diamond & Verrecchia, 1991;Healy & Palepu, 2001;Botosan, 1997).
This study is significant since it provides a major extension to already existing literature on voluntary disclosure by way of measuring the variations in the types of voluntary details disclosed by listed operating banks in Ghana on the aftermath of the banking sector rationalisation by the central bank which resulted in the collapse of the sixteen (16) banks.

Overview of Bank of Ghana
Bank of Ghana (BOG), owned entirely by the Ghana Government executes its functions under the Bank of Ghana Act, 2002 (Act 612) (as amended by the Bank of Ghana (Amendment) Act, 2016 (Act 918)), Banks and Specialised Deposit-Taking Institution Act, 2016 (Act 930) and the Public Financial Management Act, 2016 (Act, 921) undertook the recapitalization of banks in 2018 and the outcome showed a total of 23 banks meeting the minimum requirement, comprising 14 foreigncontrolled banks and 9 domesticallycontrolled banks. Whereas several stakeholders believe that the measures taken by BOG have reduced the confidence stakeholders have in the banking sector, other stakeholders have different views.

The Ghana Stock Exchange (GSE)
The Ghana Stock Exchange executes its functions under Ghana's Companies Code, 1963 (Act 179). The objectives and functions the Ghana Stock Exchange among others are to set up the space and structure for the public to buy and sell securities.
Theories influencing voluntary disclosure in corporate reporting are agency theory, capital need theory and signalling theory.
The agency theory is relevant to the study since corporate management are required to disclose adequate and relevant information relating to financial, strategy and non-financial in corporate reporting for shareholders and other accounting information users to ensure swift reduction of information asymmetry problem.

Agency theory
Agency theory enumerates the relationship between agent and principal. The corporate management (agent) are working on shareholders (principal) behalf and must therefore ensure that their actions are on the side of shareholders and the firm (Jensen & Meckling 1976;Morris 1987). Agency problem exist where corporate management are inclined to execute actions in their self-interests inconsistent with that of shareholders leading to information asymmetry (Jensen & Meckling 1976). Adequate disclosures on financial, strategy and non-monetary information in corporate reporting will therefore serve as a measure to control the activities of corporate management and lessen information asymmetry problem as well as agency cost (Ball & Foster, 1982;Jensen & Meckling, 1976;Morris, 1987;Barako, Hancock & Izan, 2006).

Capital need theory
It is believed that companies get listed on the stock market because of their desire to raise capital and also have the advantage of boosting their public image. Available data on the website of the Ghana Stock Exchange (2019)  Capital need theory is concerned with firms' motivation to raise capital (either equity or debt) at a low cost and mentions that, a firm's wish to raise capital from the capital market would encourage it to make adequate voluntary disclosure in corporate reporting (Choi 1973;Cooke 1989a). Voluntary disclosure reduces capital cost and enhances shares purchase. For this reason, corporate entities must see the need Meek et. al (1995) included 27 nonfinancial information whilst de la Bruslerie and Gabteni (2010) included 3 non-financial information items in their study. UNCTAD (2017) on the other hand included 6 nonfinancial information items in the case study of Itaú Unibanco.
Meek,  in their study developed 85 voluntary disclosure items to include 23 strategic information items whilst de la Bruslerie and Gabteni (2010) included 22 strategic information in their study. United Nations Conference on Trade and Development (UNCTAD, 2017) case study of Itaú Unibanco also provided 14 strategic information.
The capital need theory is relevant to this study because corporate management of listed banks operating in Ghana have the desire to raise money from capital market. As such, there is the need for corporate management to provide relevant voluntary details in corporate reporting to attract potential shareholders.
Signalling theory in corporate reporting explains voluntary disclosure as a condition to reduce the problem of information asymmetry (Ross, 1977). Companies competing for capital from investors lean on adequate disclosure of voluntary information in corporate reporting as a way of signalling investors and other users of their excellence, comparable to other companies to attract investors (Verrecchia, 1983;Campbell, Shrives & Saager, 2001). The problem of information asymmetry is reduced when corporate management decide not to hide specific information needed by users.

8. Non-financial details
Non-financial details used in view of this context is concerned with any information provided in corporate reporting discounting financial details that are not quantified in monetary terms. Disclosure of non-financial detail is relevant to voluntary disclosure studies because it is believed to provide transparent and more interesting information in corporate reporting.

Financial details
Financial detail used in this context is concerned with information such as firm's The signalling theory is relevant to the study since corporate management of listed banks operating in Ghana wish to enhance their company's credibility. As such they would signal investors and other accounting information users by way of providing adequate voluntary details in their reporting with regards to their excellence.
to disclose adequate voluntary details in t h e i r r e p o r t i n g t o e n t i c e p o t e n t i a l shareholders (Choi, 1973;Firth, 1980;Healy, Hutton & Palepu, 1999;Diamond & Verrecchia, 1991;Botosan, 1997).
Companies provide strategic details in corporate reporting to disclose what they seek to achieve in the short run and long run. Strategic details have become relevant in voluntary disclosure studies since they furnish potential shareholders and other accounting information users with details concerning the long-term planning, directions of the firm and information needed for a firm to achieve its business objectives.
Whilst Copeland and Fredericks (1968) used 6 specific indices of 200 NYSE firms for the year 1964, Singhvi and Desai (1971) in their study selected 34 voluntary disclosure items, Buzby (1975) develop an index of 39 items of financial and nonfinancial details for 44 NYSE and American Express Company (AMEX) firms plus 44 th OTC firms annual reports (from 30 June, th 1970 to 30 June, 1971). Firth (1979 and1980) however developed 48 items for 40 non-listed producing entities and 100 listed producing entities' annual reports for the year 1976 and Chow and Wong-Boren (1987) developed 24 items for the annual reports of 52 Mexican Stock Exchange firms for 1982 period. Cooke (1989a and1989b) on the other hand developed 224 items and 146 items respectively for 90 entities in Sweden for the 1985 period and confirms that 'listing status and size' explain the extent of disclosure and also finds that 'listing status and size' are important explanatory variables for voluntary disclosure whereas trading entities disclose little voluntary details as compared to industrial entities. assets, liabilities, equity that can be found in the annual report. Financial details are essential to disclosure studies since they p r o v i d e i n f o r m a t i o n o n fi n a n c i a l capabilities of corporate entities for potential shareholders and other accounting information users.
Meek et. al (1995) included 35 information items in their study whilst de la Bruslerie and Gabteni (2010) included 15 financial information items in their study. UNCTAD (2017) however included only value added statement as financial information in the case study of Itaú Unibanco.
Researchers like Copeland and Fredericks (1968), Singhvi and Desai (1971), Buzby (1975), Firth (19791980), Chow and Wong-Boren (1987), Cooke (1989a) were inspired by Cerf's work and adopted his ideas in their research because they found his technique useful. Gray, Meek and Roberts (1995) developed 128 items for 116 U.S., 64 U.K., Multinational Corporation (MNCs) and 100 Continental European multinational entities for the year 1989 find a substantial variation i n c o r p o r a t e r e p o r t i n g b e t w e e n 'internationally listed and domestically listed firms.' Meek et. al. (1995)   Several prior studies and current studies researchers have extensively examined the literature on disclosure. However, these studies fall short to address the level of voluntary disclosure of listed firms operating in Ghana as well as research on the variations in the types of voluntary details disclosed by listed banks operating in Botosan (1997) sampled 122 producing entities and examined their 1990 annual reports and confirms that enhanced disclosure is related to 'lower cost of equity capital'. However, the study reveals no evidence of relationship between disclosure level and 'cost of equity capital'. Botosan and Plumlee (2002) used 3,623 observations involved in AIMR Reports from 1985/86 and 1995/96 periods. Their study indicates that managers of entities that produce enhanced disclosure in the annual report benefit as result of 'lower cost of equity capital' and that timely improved disclosure is related to 'higher cost of equity capital'. However, the study reveals no relationship between the 'cost of equity capital and the level of investor relations' activities.
M u r c i a a n d S a n t o s ( 2 0 1 2 ) developed 92 voluntary items and presented 3 disclosure models: 'economic, socioenvironmental and total (economic and socio-environmental)' for the annual reports of 100 biggest non-financial entities listed in the 'Bolsa de Valores de São Paulo ('Brazilian Securities, Commodities, and Futures exchange -BOVESPA)' for the period 2006-2008 and find that profitability is important in the economic model and in the 'total model' but 'leverage and auditing firm' are only important in the 'economic disclosure model'. However, 'Size, Governance, Stock Issuing, Growth Opportunities and Concentration of Control' are insubstantial to the 3 models.
The study employed quantitative research design in gathering and analysing the data. As indicated earlier, there are 23 banks in Ghana currently comprising of 14 f o r e i g n -c o n t r o l l e d b a n k s a n d 9 domestically-controlled banks. Out of the 23 banks only 9 banks are listed on the Ghana Stock Exchange but only 8 are operational in Ghana. Purposive sampling was used in selecting all the 8 banks listed on the Ghana Stock Exchange and operating in Ghana. Listed banks operating in Ghana were chosen because of their desire to raise capital for their businesses from the stock market and also because of their advantage to boost their public image by getting listed on the stock market. It is then expected that, listed banks operating in Ghana disclose adequate voluntary details in their annual report to attract potential shareholders. S e v e r a l r e s e a r c h e r s h a v e developed their own voluntary disclosure index to assess the level of disclosure made by firms using either weighted disclosure indexes or unweighted disclosure index since there is currently no universally accepted voluntary disclosure index. Different weights are applied to the same item of information for all firms under weighted disclosure whereas unweighted disclosure approach use either dichotomous approach or modified dichotomous approach to attach different weights to applicable item of information for all firms. Several researchers have criticized the use of weighted index approach because certain information relevant to some companies might not be applicable and relevant to other companies. In such case, it would not be appropriate to use the same weight for all companies (Cooke 1992). This study however adopted the modified dichotomous approach under unweighted index by Cooke (1989a;1989b since it is appropriate to this study because Chow and Wong-Boren (1987) opined that users get the chance to make an independent analysis of this method and also require users to read the corporate annual reports wholly to overcome any potential bias. Cooke (1989a; 1 9 9 1 a n d 1 9 9 2 ) a l s o a rg u e d t h a t , The annual reports obtained from the various companies' website for the year 2018 were assessed to assist in this study to determine the level of voluntary details disclosed by the banks. The annual report for the year 2018 was selected because it is fairly current and relevant to this study on the aftermath of the banking industry rationalisation.

Methodology
Ghana on the aftermath of the central bank's rationalisation of the banking industry. This study therefore fills this gap by examining the level of voluntary disclosure of listed banks operating in Ghana in the annual reports for 2018 accounting year and advance to determine whether there are no significance variations in the types of voluntary details disclosed by listed banks operating in Ghana. unweighted disclosure index approach is appropriate since it does not use the same weight for all companies and reduces the impact of the subjective weights among the various user groups.
The disclosure index developed in previous research consists of a variety of items that can be used to provide scores to influence economic decision needs of users.
A d i s c l o s u r e c h e c k l i s t w a s developed based on prior studies (such as Cooke, 1989a), current studies (such as Meek et. al., 1995;Botosan, 1997;de la Bruslerie and Gabteni, 2010), and an examination of international trends and standard reporting practices (such as UNCTAD, 2017) and also based on guarded consideration of entity's complete annual report consisting of financial statements, information on corporate governance, i n f o r m a t i o n o n b u s i n e s s r e v i e w, shareholders information and corporate information among other. The disclosure checklist of 118 voluntary disclosure items (categorized into 3 namely strategic details, non-financial details and financial details) have been developed for this study considering the interest of the various user groups. Several researchers (such as Cooke 1992; Marston and Shrives 1991 etc.) have posited that the development of disclosure index can be subjective since annual reports have numerous user groups with different needs. For instance investors are concerned with information in relation to profitability and earnings rate of return while employees are concerned with information in relation to workplace conditions and compensation packages. This study has therefore made an extensive effort to reduce subjectivity by developing an index of disclosure items that are relevant to all range of users of accounting information. This is due to the fact that users' information needs are broad and some specific information meant for employees or any other user might also be relevant to potential shareholders and other users. The strategic details which disclose information on what the entities seek to achieve in the short run and long run is represented by general corporate details (25 items developed), corporate strategy (17 i t e m s d e v e l o p e d ) , r e s e a r c h a n d development (3 items developed) and future prospects (8 items developed). The nonfinancial details which disclose information that are not quantified in monetary terms is represented by details on directors (7 items developed), details on employees (13 items developed) and social policy details (6 items developed). The financial details which disclose information such as firm's assets, liabilities, equity in the annual report is represented by segment details (17 items developed), financial review (14 items developed), foreign currency information (3 items developed) and stock price details (5 items developed).
The voluntary disclosure index (VDI) unweighted index (modified dichotomous approach) is calculated by dividing overall voluntary disclosure items a firm disclose by number of applicable voluntary disclosure items.
This study assessed the various listed banks using the researcher's own scoring guidelines for the means as follows; 0.80 and above = excellent, 0.70 -0.79 = above average, 0.60 -0.69 = slightly above average, 0.50 -0.59 = average, 0. 40 -0.49 = below average and 0.39 and below = poor. VDI = Actual disclosure by a firm / Total possible disclosure. This is shown as; Where,

1. Hypothesis
This study assumes that there are no significant variations in the type of voluntary details disclosed by listed banks operating in Ghana. To confirm this, one way ANOVA was performed to test hypothesis of no significant differences in the type of information voluntarily disclosed by listed banks in Ghana as  used ANOVA for the purposes of statistical analysis in their study. Before onew a y A N O VA w a s p e r f o r m e d , t h e assumption of normality was evaluated using Kolmogorov-Smirnov test to check for normality of the three disclosure groups. Also, Levene's test was used to test for homogeneity of variances to ensure variances for the three groups are the same for the test to give a valid result. Eta squared was computed (Eta squared = Sum of squares between groups/Total sum of squares) to determine the effect size using Cohen (1988) guidelines of 0.01 = less effect, 0.06 = intermediate effect and 0.14 = great effect. To evaluate the nature of the differences between the 3 groups made up of strategic details, non-financial details and financial details, the Tukey post hoc test was used to measure the statistically significant differences between the 3 groups. The ANOVA model is specified as: Source: Authors computation from 2018 annual reports. Table 2  Bank Ltd did acknowledge the importance of providing adequate financial details in their annual reports. The implication is that, these banks admitted the important role a d e q u a t e fi n a n c i a l d e t a i l s p l a y i n c o m m u n i c a t i n g a fi r m ' s fi n a n c i a l performance to potential shareholders and other accounting information users, which essentially would boost the confidence of potential shareholders and other accounting information users.
Non-financial details in table 2 revealed that Access Bank Ltd and Standard Chartered Bank Ghana Ltd had the highest level of disclosure with means of 0.8462 each. This outcome indicates that Access Bank Ltd and Standard Chartered Bank Ghana Ltd were more concerned with nonfinancial details as compared to the other listed banks. This also implies that Access Bank Ltd and Standard Chartered Bank Ghana Ltd were more transparent in their reporting to restore confidence in investors and other users of accounting information. Ecobank Ghana Ltd followed with mean of 0.6538. Next was CAL Bank Ltd with mean of 0.6154. Ecobank Ghana Ltd and CAL Bank Ltd means are slightly above average per the researcher's level of voluntary disclosure scoring guidelines. Next were GCB Bank Ltd and Societe General Bank Ltd with means of 0.  Choi, 1973;Firth, 1980;Cooke, 1989a;Healy et al., 1999;Diamond and Verrecchia, 1991;Botosan, 1997 and also consistent with the agency theory, capital need theory and signalling theory.
Test for homogeneity of variance in table 5 revealed that the significance value (Sig. = 0.124) is greater than 0.05. This is an indication that the variances in scores are the same for each of the three groups made up of strategic details, non-financial details and financial details. This means the ANOVA is considered robust since the requirement of homogeneity has been met.
Descriptive statistics in Table 6 revealed a numerical difference between the means for the 3 groups (that is M=0.6534 for strategic details, M=0.6490 for nonfinancial details and M=0.5833 for financial details). This suggests that listed banks in Ghana disclosed more information under strategic details as compare to non-financial and financial details for the 2018 period. This is interesting to assume that, listed banks operating in Ghana had a big craving for strategic details and demonstrated that strategic details were more appealing to restore confidence in potential shareholders and other accounting information users than non-financial and financial information. Listed banks operating in Ghana on the other hand disclosed more information under nonfinancial details than financial details. This is also right to assume that, listed banks operating in Ghana had an intense craving for non-financial details than financial details and gave credence to the fact that non-financial details included in their 2018 annual reports were more attractive to potential shareholders and other accounting information users than financial details. Though there were numerical differences between the means for the 3 groups, there is no enough evidence to conclude that the differences in means for the type of information voluntarily disclosed by listed banks in Ghana are not significant. Therefore, one way ANOVA was performed to test hypothesis of no significant differences in the type of information voluntarily disclosed by listed banks in Ghana.
The Tukey post hoc test in table 8 revealed a statistical significant difference between strategic details and non-financial details with p value of 0.014 (i.e., p = 0.014) which is below 0.05. This suggests a measurable difference between strategic details and Non-financial details. This shows that the difference in mean for strategic details and Non-financial details was substantial to conclude that, the difference was very unlikely to be due to chance. This confirms the assumption that listed banks operating in Ghana had a big c r a v i n g f o r s t r a t e g i c d e t a i l s a n d 14 Vol.6.No.1 2020 International Journal of Accounting & Business Finance ANOVA table is presented in table 7. The F test of 6.377 in the ANOVA table revealed that at least one of the independent variables has a significant difference. The significance value is 0.003 (i.e., p = 0.003), which is below 0.05 shows a statistical significant difference between strategic details, non-financial details and financial details. The effect size associated with statistically significance difference between groups calculated using Eta squared shows an effect size of 0.130 which indicates an intermediate effect according to Cohen (1988) guidelines. Therefore, the null hypothesis of no significant differences in the type of voluntary details disclosed by listed banks operating in Ghana is rejected. Since the ANOVA table did not show which specific groups differed, the Tukey post hoc test was used to measure the statistically significant differences between strategic details, non-financial details and financial details.
demonstrated that strategic details were more appealing to restore confidence in investors and other users of accounting information than non-financial details in their 2018 annual reports.
There is also a statistical significant difference between strategic information and financial information with p value of 0.005 (i.e., p = 0.005) which is also below 0.05. This shows that the difference in mean between strategic details and financial details is appreciable to conclude that, the difference was very unlikely to be due to chance. This confirms the assumption that that listed banks in Ghana had an intense craving for strategic details than financial details and gave credence to the fact that strategic details included in their 2018 annual reports were more attractive to potential shareholders and other users of accounting information than financial details. However, there is no statistical significant difference between non-financial details and financial details since the p value of 0.993 (i.e., p = 0.993) is greater than 0.05.

Conclusion
The study assessed the level of voluntary details disclosed by listed banks operating in Ghana in their annual reports for the 2018 period and also determined the variations in the types of information voluntarily disclosed by listed banks in Ghana. The overall total voluntary disclosure level mean of 0.6294 is an indication that voluntary disclosure level among listed banks operating in Ghana was slightly above average (per the researcher's level of voluntary disclosure guidelines). Access Bank Ghana Ltd made the highest level of overall voluntary disclosure with mean of 0.7203 followed by Standard Chartered Bank Ghana Ltd with mean of 0.7119. Ecobank Ghana Ltd followed with mean of 0.6525, then CAL Bank Ltd with mean of 0.6154. GCB bank Ltd was next with mean of 0.6068, followed by Agricultural Development Bank with mean of 0.5862. Republic Bank Ltd was next with mean of 0.5812, and followed by Societe General Bank Ltd with mean of 0.5593. Access Bank Ghana Ltd had the highest voluntary disclosure with mean of 0.8113 for the strategic details followed by Standard Chartered Bank Ghana Ltd with mean of 0.7170. Access Bank Ghana Ltd and Standard Chartered Bank Ghana Ltd means were 0.8462 each for non-financial details. However CAL Bank Ltd had the highest mean of 0.6667 for financial details followed by GCB Bank Ltd and Standard Chartered Bank Ghana Ltd with means of 0.6154 each. Agricultural Development Bank and Ecobank Ghana Ltd followed next with means of 0.5897 each. Republic Bank Ghana Ltd and Societe General Bank Ltd were next with means of 0.5385 each. Access Bank Ltd then followed with mean of 0.5128. The results for Standard Chartered Bank Ghana Ltd which indicated mean of 0.717 for strategic details, 0.8462 for non-financial details and 0.6154 for financial details may suggest that Standard Chartered Bank Ghana Ltd disclosed more information with regards to investor's information needs than Access Bank Ghana Ltd and other listed bank. All the 8 listed banks operating in Ghana had shown their desire to raise capital from the capital market by disclosing voluntary details in their 2018 annual reports.
The Tukey post hoc test showed a statistical significant difference between strategic details and non-financial details Barako, D. G., Hancock, P., & Izan, H. Y.