Financial transparency, trust and willingness to pay in local governments of sub-Saharan Africa

Abstract

Purpose

The study investigates the influence of financial transparency on citizens' trust and revenue paying behaviour of citizens of local governments in sub-Saharan Africa. It relies on the theories of stewardship and public choice in explaining the relationship between financial transparency, trust and willingness to pay.

Design/methodology/approach

The study applied a Partial Least Square Structural Equation Model (PLS-SEM) to survey data of 404 respondents selected from four Metropolitan and Municipal Assemblies of Ghana to test the hypotheses of the study.

Findings

It establishes the fact that financial transparency positively influences trust of citizens in local government and their willingness to pay taxes and levies. The study also found that both financial transparency and trust are low in the local governments of Ghana.

Practical implications

The study emphasises the importance of financial transparency in improving trust and willingness to pay. Thus, local governments are encouraged to seek innovative ways to enhance the quality and access to financial information by the citizens.

Originality/value

While prior studies focus on the measurement and determinant of financial transparency, this study links financial transparency to revenue mobilisation in the local government of sub-Saharan Africa.

Citation

Krah, R. and Mertens, G. (2023), "Financial transparency, trust and willingness to pay in local governments of sub-Saharan Africa", Journal of Public Budgeting, Accounting & Financial Management, Vol. 35 No. 6, pp. 100-120. https://doi.org/10.1108/JPBAFM-06-2022-0110

1. Introduction

Transparency has become a universal norm of governance in the 21st century and governments are admonished to be transparent with their citizens. In public governance, transparency means allowing the public to gain information about the decisions and operations of the government (Etzioni, 2010). It is not only a means to an end but an end by itself (Birkinshaw, 2006). As a means, transparency promotes democratic participation, builds trust, disinfects corruption and permits informed decision making (Cuillier and Piotrowski, 2009; Grimmelikhuijsen, 2012; Kim and Lee, 2019). Transparency is an end in itself as it is the right of citizens to know about the happenings within government. Some governments have voluntarily adopted transparency as a way of improving organisational performance and operational efficiency (Cucciniello et al., 2017; Kim and Lee, 2019). The transparency policy of a government is shaped by the cultural milieu made up of social attitudes towards the essence of information, the level of participation of stakeholders in government, viability of independent media and freedom of information enactments (Brown and Cloke, 2004). Governments, in recent times, have established varied measures to promote transparency across the public sector including open government data initiatives, passage of freedom of information (FOI) laws and e-governance initiatives (Grimmelikhuijsen, 2010; Oztoprak and Ruijer, 2016; da Cruz et al., 2016; Kim and Lee, 2019).

Scholars have underscored the potency of transparency to improve operational efficiency and performance of organisations. Hood (2006) and Cucciniello et al. (2017) show that government transparency leads to improved financial efficiency and accountability, enabled by the improved monitoring and evaluation of organisational operations by the stakeholders. de Boer and Eshuis (2018) investigate the relationship between government transparency and regulatory performance and conclude that transparency has a positive influence on regulatory performance of inspectors of government.

The uniqueness of local government lies in the closer link with the local citizens in terms of policy and decision making at the local level. They are configured to engender increased participation in the policy making and assessment of accountability (Dowley, 2006; Grimmelikhuijsen, 2010). Transparency at the local government level is necessary for fostering the needed cooperation of the citizens for the overall achievement of local service delivery (Armstrong, 2011). A responsibility of the local government is to raise sufficient “own revenues” from the citizens. In Ghana for example, the Local Governance Act, 2016 (Act 936 as amended) specifically entrusted the local government with the responsibility of formulation and execution of plans, programmes and strategies for the effective mobilisation of resources necessary for the overall development of the municipality. Consistent with Niskanen's budget–maximizing hypothesis (Niskanen, 1971), local managers aim at maximizing the budget of the local government in order to increase the resources available for the provision of public services. How effectively a local government maximizes the budget depend partly on the willingness of the citizens to pay. The inability of the local government in developing countries to raise the needed resources continue to undermine the delivery of public services (Fjeldstad and Heggstad, 2012; Adu-Gyamfi, 2014). There are known causes of low revenue mobilisation in local governments, such as revenue leakages through corrupt practices of revenue collectors, weak revenue management systems and lack of enforcement of compliance (Fjeldstad and Heggstad, 2012; Adu-Gyamfi, 2014; Mgonja and Poncian, 2019). There is evidence of a growing unwillingness of the citizens to pay taxes and fees (Puopiel and Chimsi, 2015; Konlan, 2015), perhaps as a protest against lack of transparency in how the resources are used. Prior studies have emphasized the role of governance effectiveness in shaping the tax paying behaviours of citizens, suggesting that an accountable and a transparent governance may positively influence citizens' willingness to honour their financial obligations to the local government (Ortega et al., 2016; Du Preez and Stoman, 2019; Iraman et al., 2021).

Porumbescu et al. (2017) assert that transparency is capable of enhancing the citizens' intention to comply with public policy, especially in areas of policy dealing with routine trade-offs relating to improvement of quality-of-life issues such as provision of quality services. Moreover, they find that understandable policy information increases policy appreciation and greater willingness to comply. In support, we argue that financial transparency of local governments may serve as an effective tool in harnessing resources with ease at least cost through greater willingness to pay.

The financial reporting in local governments, like all other organisations, is mandatory but most often carried out in compliance with the laws without cognisance of the budget maximisation effect. Porumbescu et al. (2017) provide an insight into how local governments may explore financial reporting as a tool to create greater understanding among citizens about how their monies are used in order to influence their willingness to pay. Currently, local governments in some advanced countries have recognised the strength of financial transparency in creating understanding and complying citizens through financial reporting initiatives. An example of such financial transparency initiative is the popular reporting initiative of local governments in the USA, which breaks down the financial information to a level that an ordinary citizen will understand how the public financial resources are employed (Yusuf et al., 2013; Manes-Rossi et al., 2019). Manes-Rossi et al. (2019) assert that transparent financial reporting to citizens creates a favourable environment for citizen's involvement in the decision-making process, leading to democratic participation.

However, little is known about how local governments in Sub-Saharan Africa are exploiting the budget maximizing effect of financial reporting. Financial transparency in local governments in Sub-Saharan Africa is reported to be very low (Otenyo and Lind, 2004; Krah and Mertens, 2020a), perhaps accounting for the poor revenue performance of the local governments. Our study argues that financial transparency is an antidote to revenue mobilization challenges of the local government in Sub-Saharan Africa, a call which hitherto was not loud enough in the public sector accounting literature. Barbera et al. (2016) call for the promotion of popular reporting initiatives everywhere because it strengthens governance and the capacity to ensure greater transparency in government. This study. Therefore, aims at linking financial transparency to the revenue mobilisation, measured by the citizens' willingness to pay in Ghanaian local governments. The objectives of the study are of two folds. First is to find out whether more transparency in financial matters leads to more trust of citizens in their local government and second is to examine whether trust of citizens translates increased willingness to pay local taxes.

The paper proceeds as follows. The second section provides a theoretical framework and literature review on transparency and accounting information. The third section describes the research methodology. Section four presents the results and section five discusses the results of the study respectively. The final section presents the implications of the study, the limitations as well as future research avenues.

2. Literature review

2.1 Transparency in local government

According to Grimmelikhuijsen (2012) transparency is about the availability of information about an organisation which allows actors outside the organisation to monitor the internal workings and performance of the organisation. It entails revelation of information about public organisation's decision processes, procedures, functioning and performance (Curtin and Meijer, 2006; Grimmelikhuijsen, 2012). Krah and Mertens (2020b) examine the trend and patterns of local government transparency in the 21st century and conclude that there is an upsurge in transparency in local governments in Europe and Americas, but not same in Africa. Further, Fox (2007) identifies two forms of information disclosure: proactive transparency and demand-driven transparency. Proactive dissemination describes information that government willingly discloses about its activities and performance whilst demand driven transparency arises from an institutional commitment to respond to information request of citizens for a specific kind of information or document, which will ordinarily not be provided by government.

This study aligns with proactive transparency by local governments where all relevant information is made accessible to the citizens without request. Yang (2018) also distinguished between actual transparency and perceived transparency. Actual transparency refers to the measure of transparency efforts of an organisation based on an explicit objective criterion. This assessment is usually made by the public organisation itself. On the other hand, perceived transparency is an impression of a public organisation's practice of information disclosure in the mind of the citizens. Citizens' perception of transparency is complex because it is affected by factors within control and outside the control of the organisation (Esaiasson, 2010). The controllable factors include the information disclosure policies and practices of the organisation while the uncontrollable factors include probable information asymmetry between the provider and the receiver of the information as well as individual and psychological differences in citizens' perceptions of transparency of the organization. Van der Cruijsen and Eijffinger (2010) observe that the way and manner in which individuals receive, interpret and respond to government disclosures differ greatly. Since this study looks at transparency from a citizens' perspective, perception of citizens is critical. Thus, transparency in this study connotes perceived transparency.

Financial transparency, a dimension of transparency, is concerned with the provision of information about financial management outcomes which include fiscal policies, budgeting, accounting and other financial information about a public organisation (Muñoz et al., 2017). It is achieved through preparation and presentation of financial statements using generally accepted accounting principles for public sector (Bastida et al., 2020). Financial information disclosed to the external users includes public revenues, spending and debt, information on budget, evaluations and audit opinion for financial oversight (Guillamón et al., 2011). Financial transparency is regarded by international organisations such as International Monetary Fund (IMF), World Bank and the International Public Sector Accounting Standards Board (IPSASB) a springboard for development of national economies and fighting corruption. IPSASB, for example, aims at promoting financial transparency in the public sector through the development of high-quality accounting standards (IPSAB, 2021).

Financial Transparency entails the provision of accounting information that serves the needs of the stakeholders. According to van Helden and Reichard (2019), accounting information represents financial information provided in budget reports, annual accounts and other user tailored documents for planning, controlling and accountability of projects and programmes. It may also include non-financial information that enhances the usefulness of the information, in terms of understandability (Kober et al., 2010). Accounting information serves as means of rendering accountability to the stakeholders in a transparent manner and also for supporting decision making (van Helden and Reichard, 2019).

The literature has identified and classified the users of the accounting information variously, including internal users and external users, primary users and other users (See for example Mack and Ryan, 2006; Mack and Ryan, 2007). Interestingly, citizens have been identified as the most important normative interest group. The IPSASB (2021) classified citizens as primary users of the accounting information whose interest is paramount in the preparation of financial reports. However, Cohen and Karatzimas (2015) observe that the citizens are often remote from their municipality and only few of them, occasionally, have access to the accounting information through mass and social media. This suggests that there exists financial information gap between the local government and the citizens which will likely affect the cooperative behaviour of the citizens. The implication is that greater transparency is required to bridge the seemingly gap in the flow of financial information to citizens.

2.2 Trust in local government

Generally, trust is a psychological state in which the intention to accept vulnerability is grounded upon positive expectations of the intentions or behaviour of another (Rousseau et al., 1998). It is measured by the extent of citizens' confidence in government to operate in the best interests of society (Cleary and Stokes, 2009). Trustworthiness of government, according to Grimmelikhuijsen (2010), is of three dimension: competence, benevolence and honesty. Whilst competence-based trust is relating to perceived capability, effectiveness, skilfulness or professionalism in making decisions of public organizations, benevolence-based trust develops out of belief of caring intentions of government action. Honesty-based trust is built upon the commitment and honesty of government. Fitzgerald and Wolak (2016) show that citizens' trust differs for local government and national government and that trust in local government is based on substantive factors related to local life experiences rather than the attributes of a national government. This justify the effort to study trust in local government context. In the local government, citizens' assessment of the government's performance towards a more transparency serve as a key determinant of trust in the local government (Kim and Lee, 2012). Thus, trust increases when the local government is perceived to be competent in delivering public service and that the managers of the local government genuinely care about the needs of the citizens by conducting their affairs in an honest manner. Often, transparency is regarded not only as panacea for better governance but also as tool for building trust in government (Hood, 2006; Grimmelikhuijsen and Meijer, 2014). Accordingly, the citizens' perceived financial transparency of the local government has implication for trust building in the local government (Grimmelikhuijsen and Meijer, 2014; Yang, 2018; Puron-Cid et al., 2019).

2.3 Theoretical underpin and hypotheses

The study chooses to combine stewardship and public choice theories to explain how budget maximizing local managers use financial transparency as a tool to stimulate citizens' trust and willingness to pay. Paramount in the local government context is the sense of stewardship, where the utility and satisfaction of the local managers emanate from achievement of success through efficient application of resources (Keay, 2017). Stewardship theory emphasises the subjugation of personal interest and short term gains for common good and altruistic achievement in trustworthy management of public resources (Laszlo et al., 2003; Contrafatto, 2014). A steward gets fulfilment for the success of the organisation and therefore stewardship behaviour is organisation-centred (Donaldson and Davis, 1991). Davis et al. (2010) posit that stewardship behaviour and preferences are aligned to long term aspirations of the organisation for the benefit of a cross section of stakeholders. It has been argued that stewardship works better in context where parties are willing to trust each other (Davis et al., 1997). In stewardship relationship, trust building is the preoccupation of the steward in order to propel organisational attainments (Keay, 2017).

Furthermore, as stewards the managers of the local governments are confronted with situation that require policy choices and decision making. Buchanan and Tollison (1984) put forward public choice theory that explains the behaviour of individual actors in the public sector, including public managers, in terms of observable public decision outcomes. It assumes that public decision makers are faced with many alternatives and preferences are influenced by the utility function of the policy maker. In policy implementation, the theory predicts that bureaucrats have a tendency to make choices that result in outcomes favourable to their own interest. In that, the bureaucrats maximize the budget of the entity beyond plausible efficiency limits so as to improve service delivery to the constituents. Niskane's (1971) budget maximization hypothesis states that public managers have the propensity to maximise the size of their budget for utility maximization and sustainability purposes to meet the minimum cost of supplying an expected output (Breton and Wintrobe, 1975; Imbeau, 2017).

This implies that a budget-maximizing steward, in an attempt to win public support, uses financial transparency as a tool to win the trust of the citizens in seeking their willingness to pay for their preferred public services. The local managers are often elected by the local citizens to seek the public interest through improved service delivery and this provides a context of stewardship relationship evidenced by public-centeredness and trust. Hence, the local managers are deemed to make policy choices that enhance the collective good of the community reflective of utility maximization of a steward. So, when public managers are confronted with a choice between financial transparency and information asymmetry, they will choose the path of transparency in order to gain public trust and support through a superior service performance.

The combination of stewardship and public choice theories provides an insightful theoretical contribution to the study. First, stewardship theory was used to explain trust as an essential element of stewardship relationship where pro-organisational behaviour is the norm. The theory supports the need for local managers (stewards) to find acceptable means, such as adoption of transparency, to build trust of citizens (principals) in the government. Second, the study uses the public choice theory to support the motivation of local managers to increase revenue of the local government. The theory recognises the inherent motivation of public managers to maximize the budget of their organisations, as a matter of public choice. Hence, the study applies the budget maximization hypothesis to explain why the local managers will choose financial transparency as a tool to increase revenue of the local government. The combine effect of the two theories inspired the mediation role of trust in the relation of financial transparency and willingness to pay.

2.4 Hypothesis development and formulation

Transparency is recognized as a tool for building trust with the citizens of local government and in turn for citizens' support and compliance (Porumbescu et al., 2017). Puron-Cid et al. (2019) demonstrate that financial transparency constitutes a public value capable of driving financial sustainability and combat corruption in government. Further evidence shows that citizens who are well informed about the importance and usefulness of spending policies are more likely to voluntarily comply with tax requirements (Manes-Rossi et al., 2019). Van der Cruijsen and Eijffinger (2010) state that perceived transparency is a powerful determinant of citizens' behaviour in government-citizen relationship. Stewardship theory explains that the foundation of a steward -principal relationship is trust, therefore the steward endevours to court the trust of the principal through pro-organisational behaviours. This presupposes that trust building is essential in creating a supportive local government and therefore the local manager engage in trust winning policies that reflect public value. As a public value, financial transparency may serve as the means through which trust of the citizens is stimulated. Thus, the first hypothesis of the study that:

H1.

Perceived financial transparency has a positive influence on citizens' trust in the government.

Public choice theory indicates that public policy making has self-interest undertone and one such manifestation is espoused by the Niskane's (1971) budget maximization hypothesis, which states that public managers aim at maximizing their budgets (resources availability) in order to enhance public service with self-interest lining. We argue that, a budget maximizing local manager is not only interested in building trust of the citizens but to be able translate it into willingness to pay more taxes to support the budget maximisation goal. So, transparency as a public value is a readily available tool for motivating the willingness of the citizens to pay, especially in a developing country context where trust in government is very low and resources for development are inadequate (Adu-Gyamfi, 2014; Mgonja and Poncian, 2019). We therefore hypothesized as follows:

H2.

Perceived financial transparency positively influences willingness of citizens to pay in the local government.

Consistent with the tenets of stewardship theory, trusting citizens are more understanding and compliance, demonstrating that trust is capable of producing supportive citizens. A positive relationship has been established between trust and willingness to pay taxes, implying that trusting citizens are more likely to honour their taxes and fees to the local government (Fjeldstad, 2004; Kim and Lee, 2012). Habibov et al. (2017) show that an increase in social trust is associated with a greater willingness to pay more taxes to improve public services. Further, Anderson (2017) finds that trust in government generally has a significant positive effect on citizens' willingness to pay taxes for public goods and services. Yang (2018) suggests that accurate and timely information about governments' operation increase trust of the citizens and promote cooperation. Thus, the study hypothesizes that:

H3.

Trusting citizens are more willing to pay local taxes to the local government.

Furthermore, stewardship theory emphasises trust as co-organisational behaviours and public choice theory highlights the aspiration of local managers to mobilise more revenues for development with self-interest motive. Meanwhile, transparency can aid the budget maximization drive of the local managers better when trust worthiness of the local government is enhanced. Thus, trust may strengthen the revenue harnessing potential of financial transparency in the local government. The fourth hypothesis is therefore stated as follows:

H4.

Trusting citizens are more willingness to pay when financial information about the local government is openly available.

The literature acknowledges the effect of demographic characteristic of citizens on trust and perception of transparency. Esaiasson (2010) argues that individual and psychological differences of citizens affect their perception of transparency and the manner in which individuals understand and respond to information disclosure differs significantly (Van der Cruijsen and Eijffinger, 2010). Citizens' characteristics are believed to affect citizen's attitude towards transparency, participation and civic engagement. Piotrowski and Van Ryzin (2007) found that demographic characteristics such as age, gender, education, religion and civic engagement characteristics affect citizen's cooperation. Thus, the study controls for the effect of age, gender, education, religiosity and civil awareness on the citizens' disposition to pay. The conceptual framework of the study is shown in Figure 1.

2.5 Overview of local government in Ghana

Ghana is the first country to gain political independence in Sub-Saharan Africa and an oasis of democracy in West Africa. Ghana is a constitutional republic with two tier governance structure: national government and local government. The local government is made up of Metropolitan, Municipal and District Assemblies (MMDAs) headed by the Chief Executive Officers appointed by the President and confirmed by the general assembly. Currently, there are 254 MMDAs in Ghana clothed with the highest legal, administrative and deliberative authority within the jurisdiction under the Local Governance Act (2016), Act 936 (previously Local Government Act, 1993; Act 462). The MMDAs are responsible for the provision of local services in the areas of basic education, public health, environmental protection and sanitation. Regarding financial management, the central government is pursuing fiscal decentralisation policy where power to raise revenue, spend and borrow is progressively ceded to the MMDAs (Antwi-Boasiako, 2010). Whilst the national government transfers some resources to the MMDAs, they have the responsibility of mobilizing revenues for the delivery of local public services and infrastructure. In fact, the national government often deploys revenue mobilization as a key performance criterion of the local government as revenue mobilisation is a cardinal mandate of MMDAs under both the 1992 Constitution of Ghana and the Local Governance Act (2016) (Act 936).

There are two main sources of revenues to MMDAs under the Local Government Act and these consist of decentralised transfers from central government and internally generated funds. The decentralised transfer refers to all financial supports granted by the central government to the MMDAs and these include the District Assemblies Common Fund, grants-in-aid and any other revenues transferred from the central government. The District Assemblies Common Fund is a statutory fund created to provide financial support for the development of the MMDAs and it is shared to all the MMDAs based on a formula approved by Parliament. Internally generated funds are own revenues sourced by the MMDAs, which include rates or local taxes, licences, fees and charges and investment income. Taxes on property and all rateable persons are the major source of internally generated revenues of the MMDAs but most difficult to collect due largely to uncooperative behaviours of payers, who are the citizens (Opoku et al., 2014; Yeboah and Andrew, 2020) Notably, the MMDAs are characterised by poor culture of financial transparency which makes information about how the revenues collected are used scarce or unavailable (Krah and Mertens, 2020a).

3. Methodology

3.1 Sample selection

Citizens have been identified as a primary user of the financial statements of local government (IPSASB, 2021) and therefore the study surveys citizens as users of the accounting information. In the design, the study considers the threshold of understandability of financial reports, which requires the user to have the ability and willingness to study the financial information (IPSASB, 2021). It follows that the citizens will only understand the financial information and form perception of transparency when they are knowledgeable about the operations of the local government and have the capability to study the financial report. As a result, the population of the study consists of citizens who have reasonable knowledge of the local governments operations and the capacity to study the financial information for assessing the level of transparency of the local government. The sample was selected randomly from four Metropolitan and Municipal Assemblies in the Greater Accra Region: Accra Municipal Assembly (AMA), Tema Metropolitan Assembly (TMA), Adentan Municipal Assembly (AdMA) and Ashaiman Municipal Assembly (AshMA). These Assemblies are selected due to their cosmopolitan characteristics and active citizen engagement with the local government. Cosmopolitan Assemblies have diverse population of citizens which is reflective of the entire population of the local government. More so, to ensure that the respondents are knowledgeable with the operations of the local government, respondents are sampled from active community-based association that interact with the local government on regular basis. This helps to ensure that the data collected is representative of the population of citizens who met the understandability criteria of the IPSASB, which include knowledgeability and readiness to study the financial report (IPSAB, 2021). The representation of the data is evidenced by the demographic information of respondents reported in Table 1. The limitation of the sampling procedure is that citizens who reside outside the cosmopolitan areas and those who are not technologically savvy to complete the online questionnaire are excluded from the sample. Nevertheless, the result of the study is relevant even if the data is not fully representative.

3.2 Data collection procedures

In order to ensure that the sample reflects the targeted population, the survey was administered on the official social media platforms of these Assemblies through an on-line questionnaire posted on these platforms. To increase participation, the posting of the questionnaire was conducted through gate keepers (administrators) who repeatedly post it on the platforms with a statement requesting the group members to complete the questionnaire. Following the suggestion of Liu et al. (2017) to improve the online response rate, the length of the questionnaire was kept short with an average completion time of three minutes. In addition, the study engaged the gate keepers to personalize the invitations and use pre-notification and reminders to whip up interest in the completion of the questionnaire (Van Mol, 2017). In order to overcome the problem of multiple response associated with online survey, the restrictive control tool imbedded in the survey application was activated to limit response to once. The questionnaire was administered within one month. In all, 404 responses were obtained and used, which is an adequate sample size for PLS-SEM technique based on the 10 times rule (Hair et al., 2011).

3.3 Characteristics of respondents

The questionnaire solicited information on the characteristic of the respondents: age, gender, education, civil awareness and religiosity. The profile of the respondents displayed in Table 4.1 shows that most of the respondents (69.06%) fall within the age of 30–50 years. This implies that most of the respondents fall within the active working group and are expected to contribute to the development of the local authority. In relation to gender, over 63% of respondents are female and this may be due to the fact that female constitutes a large proportion of the population. The majority of the respondents (86%) have tertiary education, which is consistent with the characteristics of the targeted population of the study in relation to the level of knowledge and capacity to study the financial report threshold. It is worthy of mentioning that over 67% of respondents have indicated that they are fully aware of their civic responsibility toward the local government whilst 24.5% confesses that they are somehow aware of their civic responsibilities. Regarding religiosity, the majority of the respondents (81.18%) also claimed to be highly religious and this is consistent with the country's statistics on religious distribution (Ghana Statistical Service, 2022).

3.4 Measurement of constructs

The study is concerned with three constructs: financial transparency, trust and willingness to pay. To ensure content reliability, the items were adapted from the existing literature, and modified to reflect the study context. Financial transparency is measured with six reflective indicators formulated around availability of financial information of the local government to the citizens (da Cruz et al., 2016; Puron-Cid et al., 2019). Trust was measured by four indicators that reflects the citizens trust that monies of the local government are managed competently and honestly (Grimmelikhuijsen, 2012; Grimmelikhuijsen and Meijer, 2014). Willingness to pay is measured with five reflective items representing the willingness of citizens to cooperate and comply with tax demands of the local government (Fjeldstad, 2006; Byaro and Kinyondo, 2020; Iraman et al., 2021). The 15 items were measured using a five-point Likert scale ranging from 1, strongly disagree, to 5, strongly agree. To ensure face validity, the instrument was pre-tested with 5 experts in local government to check whether the instrument measures the concepts intended and the result was affirmative. Thereafter, the pilot test was carried out with 30 students reading Masters in Public Administration programme to assess the instrument in terms of wording, clarity, relevance and time spent. The pilot result confirms that the questionnaire is comprehensible and simple to complete within the required time, except for some slight changes that were recommended and effected accordingly.

3.5 Data analysis technique

The study used a Structural Equation Modelling (SEM) technique in unfolding the relationship between the constructs not only for its usefulness in evaluating complex model relationships simultaneously but also for its flexibility (Wong, 2013). Specifically, the partial least square (PLS)-SEM technique was used to test the structural relationship between the constructs. The PLS-SEM was chosen because it accommodates both formative and reflective constructs, normal and abnormal data distribution (Wong, 2013). PLS-SEM uses composite model which accommodates common, specific and error variance from the independent variables that assist in predicting the variances in the dependent variables (Hair et al., 2017). Consequently, the composite model is chosen over the factor model because it is effective in maximizing variance explained in trust and willingness to pay. The study follows the two steps involved in deploying SEM as stated by Hair et al. (2014): testing the reliability and validity of the measurement models and assessment of the structural model to test hypotheses.

4. Results

4.1 Descriptive statistics

The summary of the descriptive statistics of the constructs are provided in Table 2. It shows that the respondents generally perceived financial transparency to be below average expectation as all the six items were scored below 2.5, which is consistent with the findings of Krah and Mertens (2020a) and Otenyo and Lind (2004), who conclude that financial transparency in local government of sub-Saharan Africa is low. Surprisingly, accessing financial information through website and social media is an alien practice in the local government as it has the lowest mean score. This is inconsistent with the global trends of increasing deployment of information technologies to promote transparency in the local government (Bearfield and Bowman, 2017; Krah and Mertens, 2020b).

Trust of respondents in the use of public money is also low, having mean score of 2.5. This implies that the respondents do not trust the local government enough to use the public resources economically, efficiently and effectively. It also shown that willingness to pay is also low, implying a high level of resistance and avoidance to pay among the respondents. These results indicating low level of financial transparency, trust and willingness to pay in the local government are consistent with the findings of Otenyo and Lind (2004), Krah and Mertens (2020a)Bratton and Gyimah-Boadi (2016) relating to the Sub-Saharan Africa.

4.2 Measurement model

The overall model assessment reveals that the model is a satisfactory fit to the data as the Q², F² and Adjusted R² shown in Table 3 are all within the recommended quality criteria values. The Q² also confirms the modest predictive power of the model as all the Q² values are greater than 0. As a rule of thumb, Q² values larger than zero are meaningful: values higher than 0.25 and 0.50 depict small, medium and large predictive accuracy of the PLS path model respectively (Stone, 1974). The F² measures the size effect of the model and Cohen (1988) suggests that the effect is large when F² is 0.35, medium when F² is 0.15 and small when F² is 0.03. The size effect of the model is large as all the F² values reported exceeded the 0.35 threshold. The Adjusted R² of Financial Transparency and Trust are 0.55 and 0.32 respectively, which is considered to have a substantial predicting power (Hair et al., 2019).

4.3 Reliability and convergent validity

The study performed a test of internal consistency of the instruments using factor loading, composite reliability and Cronbach alpha. The outcomes of the factor loadings in Table 4.3 show that the items were good measures of their respective constructs, as the factor loadings exceeded the 0.5 minimum threshold of convergent validity (Byrne, 2016). A composite reliability of at least 0.70 is recommended to ensure internal consistency of the measurement (Sarstedt et al., 2014) and, as indicated in Table 3, it exceeds the threshold for every indictor and therefore acceptable. The Cronbach Alpha, further, confirms the internal consistency of the measurement as they all exceeded the threshold of 0.70 (Fornell and Larcker, 1981). Again, the model was tested for collinearity using Variance Inflation Factor (VIF) to ensure that coefficient estimates are efficient. VIF values shown in Table 3 met the criteria, demonstrating that data suffers from no multicollinearity challenge (see Hair et al., 2017). In testing for validity, the study relied on Average Variance Extracted (AVE) to measure convergent validity, and Fornell and Lacker criteria to measure discriminant validity. The AVE reported in Table 4.3 are acceptable as all exceeded the threshold of 0.50 (Hair et al., 2014), indicating that the constructs explain more than 50% of the variance of its items. The results in Table 4 shows that discriminant validity is achieved, as the correlations among the various constructs are less than the square root of the AVE scores for the constructs (see Fornell and Larcker, 1981).

4.4 Analysis of structural model

The result of hypothesis testing is shown in Figure 2 and Table 5. It is revealed that trust of citizens in the local government is significantly improved by perceived financial transparency of the local government, thus supporting H1 (H1: β = 0.56, Std dev = 0.038, p < 0.000). The result also shows that willingness to pay taxes is significantly and positively influenced by perceived financial transparency, which support the H2 (H2: β = 0.43, Std dev = 0.055, p < 0.000). H3 is about the relationship between Trust and the Willingness to Pay and the results support the claim that willingness to pay is significantly and positively influenced by trust of the citizens (H3: β = 0.36, Std. dev. = 0.056, p < 0.000). The result also suggests that the interaction of financial transparency and trust produces significant increase in willingness to pay, thus the indirect influence of financial transparency on the willingness to pay through trust is positive and significant. This is in support of H4 (H4: β = 0.20, Std dev = 0.030, p < 0.000). In relation to the control variables, the study establishes that only civil awareness and educational of respondents positively affect the willingness to pay. Age, gender and religiosity were found not to significantly affect willingness to pay, nevertheless the direction is indicative that men are less willing to pay than women.

5. Discussions

The first hypothesis (H1) states that financial transparency positively influences trust of the citizens, which is confirmed by the result. Therefore, H1 is accepted and the conclusion is that financial transparency builds the trust of citizens and the lack of it brings forth to mistrust. The result is consistent with the stewardship theory which emphasises the positive influence of pro-organisational behaviour of the steward (such as being transparent) on the trust of the principal. The findings demonstrated that when financial affairs of the local government are conducted openly, citizens are more likely to trust the local managers. The finding of the study does not only support the position of prior studies but contradict some as well. It is consistent with the finding of Alessandro et al. (2021) which emphasized that increasing transparency improves trust in the government and that the relationship between transparency and trust may be mediated by the performance of the government. Hood (2006), a strong proponent of culture of openness, believes that the perception of having an open culture has a positive effect on trust of citizens for government. Kim and Lee (2012) also reported a positive association between e-participants’ assessment of government transparency and their trust in the local government. Song and Lee (2016) investigate the use of social media, perceived transparency and trust in government and conclude that the use of social media enhances perceived transparency of the government which increases citizens trust in the government.

The finding also has contradiction in the literature. Grimmelikhuijsen et al. (2013), for example, in a cross-national comparative experiment conclude that transparency has a negative effect or no effect on trust in government. Worthy (2013) examined the connection between transparency and trust in local government in the United Kingdom and concludes that transparency has no systematic effect on trust as the medium through which citizens access the information about government may influence their level of trust. However, a more nuanced analysis of Grimmelikhuijsen and Meijer (2014) brings further clarity to the conflicting result on the relationship between transparency and trust. In their study, they observed that different categories of participants have different perception of transparency and trust. They reported positive significant effect of transparency on perceived benevolence trust among participants with relatively low levels of prior knowledge about a policy topic and who have a rather low predisposition to trust government in general. Low transparency is found to decrease perceived competence-based trust of government among participants with a rather positive predisposition toward government and limited prior knowledge about a certain policy. Again, no relationship is established between transparency and trust among participants with high knowledge, since they have already formed their opinion about local government. The result of our study is consistent with the outcomes of the first and second categories of participants in the experiment of Grimmelikhuijsen and Meijer (2014). It suggests that citizens with low levels of prior knowledge about a policy topic are likely to develop benevolent trust and competent-based trust for the government when disclosed with information about the policy. Grimmelikhuijsen (2012) demonstrates that the connection between transparency and trust in a governmental organization is much more complex: a mix of knowledge and feelings of citizens.

The second hypothesis (H2) is that financial transparency influences the willingness to pay in the local government. H2 is accepted, concluding that financial transparency positively influences willingness to pay, which means that where citizens perceived the local government to be transparent in financial matters, they tend to be more willing to pay taxes and levies imposed on them for development. The finding is in line with the public choice theory, which posits that local managers desire to maximize the budget of the local government as a matter of public choice and therefore will adopt policies that are consistent with the goal. In this case, a local managers adopt financial transparency mainly because it assists them to increase revenues of the local government.

This conclusion is also consistent with that of Porumbescu et al. (2017) who find that transparency is an enabler of the citizen's intention to comply with public policy, especially when such policy promises to improve quality of life. This implies that transparency of local government does not only create complying citizens but also understanding and supportive citizens. The claim also syncs with Manes-Rossi et al. (2019) who conclude that citizens who are well informed about how public monies are spent are often better prepared to support financial sustainability through voluntary tax paying behaviours. Similarly, Capasso et al. (2021) demonstrate with evidence that increasing access to information about government fiscal accounts and financial transactions increases citizens' willingness to pay taxes. So, local governments that are more transparent in the disclosure of their financial transactions are more likely to have citizens who are more willing to pay their taxes and other levies.

The third hypothesis (H3) states that perceived trust has a positive influence on willingness to pay, which has been accepted. The conclusion is that perceived trust and willingness to pay are positively related in the local government. This finding is consistent with stewardship theory and the existing empirical literature. According to the stewardship theory, when a steward behaves in a manner that promote the general wellbeing of the organisation, trust of the principal increases, leading to supportive conduct. Consequently, stewardship theory adequately explains the finding of the study that once trust is built, the citizens are more likely to support and voluntarily comply with policies and initiative taken by the local managers for the collective good.

Anderson (2017), in a study of the effect of trust on willingness to pay taxes for public goods and services in transition countries, concludes that trust in government generally has a significant positive effect on willingness to pay taxes for public goods and services, caveating that variation exist among government institutions and public goods concerned. Oh and Hong (2012) examine whether citizens' trust in government affects their willingness-to-pay and found a positive association between the two variables. In Tanzania, Byaro and Kinyondo (2020) also shows that higher level of citizens' trust in government encourages willingness to pay taxes in order to improve public services.

The final hypothesis (H4) postulates that the relationship between financial transparency and willingness to pay is stronger when the trust exists. The evidence support the claim and therefore H4 is accepted, concluding that trust mediates the relationship between financial transparency and willingness to pay. This demonstrates that financial transparency boosts trust of citizens which results in greater willingness to pay taxes and other levies due to the local government. The finding is explained by the combined effect of stewardship theory and public choice theory in the study. Whilst steward theory explains the trust building effect of transparency, the public choice theories justifies the end game, which is budget maximization. This suggest that a self-interest seeking manager will not only use transparency to build trust but will ensure that it culminate in increase revenue mobilisation for the local government. The study's position on the mediation power of the trust on the relationship between financial transparency and willingness to pay is also consistent with the empirical literature that suggest that transparency improves willingness to pay through trust building (Oh and Hong, 2012; Grimmelikhuijsen, 2012; Worthy, 2013; Byaro and Kinyondo, 2020).

In connection with the control variables, the study finds that education and civil awareness of respondent positively influence the willingness to pay. This agrees with Piotrowski and Van Ryzin (2007), which studied the attitude of citizens towards transparency and concluded that educational level and level of civil awareness of the citizens positively affect their support for the local government.

6. Implications of the study

The results of this study have theoretical, policy and practical implications. The study constructs the relationship between citizens and the managers of the local government using stewardship theory and public choice theory. Applying the stewardship theory, the study demonstrates that creation of a trustworthy government is a primary aspiration of local managers and therefore they undertake initiatives and policies that build citizens' trust. As a result, financial transparency is a tool employed by local government to foster trust of the citizens in the local managers to use public finances more effectively and efficiently. Relying on public choice theory, the study provides evidence to the effect that when local managers are faced with choice between financial transparency and asymmetry, they will choose transparency in order to achieve their self-interest of maximizing their budget, which aligns with public choice. The study shows that a budget maximizing local managers choose to increase revenue mobilisation so as to be able to deliver more public services to meet the expectation of the citizens. Therefore, the positive relationship, reported in the study, between financial transparency, trust and willingness to pay is a confirmation of the stewardship theory and public choice theory in local government in Ghana.

It is clear that financial transparency is key in building trust in government among the citizens and courting their willingness to pay their taxes and other levies which in turn maximizes the budget of the authority. The local government should therefore pay attention to the preparation, presentation and dissemination of quality financial information to the citizens in a timely and engaging manner. They may take a cue from the popular financial reporting initiatives elsewhere and consider the ways of making the financial information very accessible and user friendly to the citizens of the local government. Such policies may make financial transparency an effective tool in budget maximization efforts of the local government. Meanwhile, transparency is not achieved only by provision of financial information to the citizens but also the citizens' ability to translate the information provided into favourable perception about the local government. Therefore, the local government should seek to understand better what will positively influence this perceived transparency. Since this is different for different cultures and countries (and even within countries on a local scale), there is no standard formula for increasing transparency and therefore local government should invest in understanding the needs and demands of the citizens so they can cater to their specific needs.

The practical essence of the study is that the local government should no longer take financial reporting as usual business but rather be concerned with the quality, understandability and access to the financial information. This implies that financial information should be prepared in compliance with the International Public Sector Accounting Standards (IPSAS) which promises to deliver quality and useful financial information to all users, including the citizens. Thus, the study calls on the local governments is Sub-Saharan Africa to adopt and implement the IPSAS to promote financial transparency. Further, the local government must recruit competent and professionally qualified accountants to perform the financial reporting function. The local government should practically explore the use of social media and websites in the provision of financial information to the citizens in this fast-growing information technology world. They should not only ensure provision of quality financial information but also ensure that the citizens are educated on how to use the financial information. In order to better understand the user (citizen) needs and demand for financial transparency, the local government could set up town hall meetings and meet with user groups to interact and identify the information needs; send out online surveys to reach out to a broader audience and determine user needs and undertake radio discussions in the various local languages to educate the citizens to become better able to understand the financial information of the local government. This will help to increase financial literary among the citizens and increase the perception of transparency of the local government. Thus, the financial reporting and making the information understandable to the citizens should be prioritised by management to ensure that financial transparency requirements are met.

7. Limitations and further research

The online survey approach used might have affected the response rate and the sample size of the study, nevertheless the sample size is proven to be statistically sufficient. It might have made it difficult or impossible for those respondents who are not technology inclined to complete the questionnaire online. A further limitation of the study is the concentration of respondents in the cosmopolitan assemblies, which perhaps limits the representation of the population. Nonetheless, the cosmopolitan assemblies are known for population diversity and drawing sample from them ensure that respondents represent a fairly broad group of citizens that use the financial statements of the local government thereby rendering the potential sampling bias is insignificant. Consequently, future studies may consider the use of survey experiment to explore the effects of financial transparency in the local government context. This may provide some useful insight into the dynamics of perceived financial transparency within the various groups of participants. In addition, the study assumes that the mere availability of financial information to the citizens translates into financial transparency without paying attention to the usefulness of the financial information to the citizens. It is recommended that further studies are conducted to examine how the local governments can provide a citizen centric financial information to deepen the perceived financial transparency among the citizens.

Figures

Figure 1

Conceptual model

Figure 2

Result model

Table 1

Profile of respondents

Characteristics	Frequency (n = 404)	Percentage (%)
Age
Less than 30 years	109	26.98
Between 30 and 50 years	279	69.06
Above 50 years	16	3.96
Gender
Male (1)	148	36.63
Female (0) Others	256 0	63.37 0.0
Religiosity
Not religious	10	2.48
Somehow religious	66	16.34
Highly religious person	328	81.18
Education
Pre-tertiary	54	13.37
Tertiary	240	59.41
Post-tertiary	110	27.22
Civic Awareness
Can't tell	34	8.42
Somehow aware of civil responsibilities	99	24.5
Fully aware of civic responsibilities	271	67.08

Source(s): Field Survey Result, 2021

Table 2

Descriptive statistics for constructs

Indicators	Obs	Mean	Std. Dev
Perceived financial transparency FT1: I am aware that financial information is provided on the Authority's web site	404	1.681	0.910
FT2: I am aware that financial information of the Authority is on the social media	404	1.604	0.922
FT3: I am aware that financial information of the Authority is available at the Office	404	1.956	1.015
FT4: I am aware that financial information of the Authority is available at the Community Centres	404	1.839	0.990
FT5: I am aware financial information of the Authority disseminated through emails	404	1.592	0.865
FT6: I am aware the financial information of the Authority is discussed on radio/TV	404	1.735	0.849
Trust in the use of public money T1: As for my Authority, I know it will use the money collected from us well	404	2.064	0.994
T2: I believe that my Authority always conduct its financial dealings well	404	2.012	0.895
T3 My Authority has competent and honest managers to apply our money well	404	2.072	0.901
T4: I believe my Authority means well for the Community	404	1.871	0.952
Willingness to Pay (WTP) WTP1: I always pay the local taxes and fees without any delay	404	2.2401	0.848
WTP2: I always pay the local taxes and fees imposed without any hesitation	404	2.1906	0.951
WTP3: I always see local taxes and fees as reasonable	404	2.0248	0.874
WTP4: I believe that the local taxes and fees are commensurate with the services offered	404	1.9653	1.008
WTP5: I do encourage people close to me to pay their local taxes and fees promptly	404	2.0470	0.851

Source(s): Survey Instrument, 2020

Table 3

Reliability and validity results

Construct	Item	Factor Loading	VIF	Cronbach's Alpha	Composite Reliability	AVE
Financial Transparency	FT1	0.883	3.525	0.953	0.958	0.677
	FT2	0.875	3.455
	FT3	0.857	2.783
	FT4	0.880	3.303
	FT5	0.881	3.238
	FT6	0.831	2.303
Trust	T1	0.861	2.329	0.891	0.924	0.753
	T2	0.904	2.910
	T3	0.866	2.381
	T4	0.840	2.131
Willingness to Pay	WTP1	0.726	1.822	0.856	0.896	0.634
	WTP2	0.838	2.463
	WTP3	0.825	2.046
	WTP4	0.801	1.858
	WTP5	0.786	1.773

Source(s): Survey Results, 2020

Table 4

Fornell and Larcker's discriminant validity

	Age	CA	Educ	FP	FT	Gender	Reli	Trust
Age	1.000
Civic Aware	0.124	1.000
Educ	0.433	0.271	1.000
WTP	0.114	0.249	0.190	0.796
FT	0.071	0.143	0.096	0.659	0.868
Gender	0.005	0.056	−0.120	−0.059	−0.105	1.000
Reli	0.030	0.115	0.013	0.133	0.110	−0.061	1.000
Trust	0.068	0.198	0.113	0.628	0.563	0.036	0.100	0.868

Source(s): Survey Results, 2020

Table 5

Result of hypothesis testing

Hypothesis	Coefficient	Std dev	P-value	Conclusion
H1: FT → T	0.56	0.038	0.000	Not rejected
H2: FT → WTP	0.43	0.055	0.000	Not rejected
H3: T → WTP	0.36	0.056	0.000	Not rejected
H4: FT → T → WTP	0.20	0.030	0.000	Not rejected