The effect of non-profit organizations’ characteristics on their transparency.

(1)

The effect of non-profit organizations’ characteristics on their transparency.

Name: Tamar de Windt Student number: 10965882

Proposal supervisor: Dr. Réka Felleg Date: 21 June 2021, Final version Word count: 9662

MSc Accountancy & Control Specialization: Control

EBEC code: 20210128110156

Faculty of Economics and Business, University of Amsterdam

(2)

Statement of Originality

This document is written by student Tamar de Windt who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

This study aims to evaluate which characteristics of the non-profit organizations plays a significant role in the transparency of the non-profit organizations. To answer the research question posed for this study and test the proposed hypotheses, an archival research method was conducted. The method used to collect data for this study is based on random sampling and is hand collected from the databases of the website company.info. I predicted that the annual report of larger non-profit organization is more transparent than smaller non-profit organizations. I document that organizational size only influence the additional effort and not to the decision to be transparent. Prior literature shows that high-quality auditing is particularly important for companies that are in need of finance (Beisland et al., 2015). For this particular reason, I tested whether the Big Four auditors add to the transparency of these non-profit organization. The results indicated that Big Four auditors do add to the transparency of non-profit organizations and enhance the credibility of their financial reports. Lastly, this study also aims to tests whether the proportion of the contributions received by donor support has a positive effect on the

transparency of the non-profit organizations. It is concluded that there is enough evidence to show that non-profit organizations that rely on donor support are more likely to provide

additional value-relevant information to ensure accountability to donors to continue generating enough funds. The ratio of contributions received only influences the additional effort, not the decision to be transparent. This study thus contributes to providing the stakeholders with a greater level of accountability and credibility from the non-profit sector and a better understanding of which characteristics influence transparency.

Keywords: Transparency, non-profit organization, disclosure

(4)

Table of content

1 Introduction 5

2 Literature review 8

2.1 Agency theory 8

2.2 Accounting transparency in corporate reporting 8

2.3 Non-profit organization 10

2.4 Voluntary disclosure for non-profit organization 11

2.5 Regulations in the Netherlands 12

2.6 Hypotheses 13

3 Research methodology 16

3.1 Sample 16

3.2 Variable measurements 17

4 Empirical results 24

4.1 Descriptive statistic 24

4.2 Regression analysis 29

4.2.1 Organizational size 29

4.2.2 Big four auditors 31

4.2.3 Revenue generation 31

4.3 Robustness analysis 32

6 Conclusion 35

Reference list 37

(5)

1. Introduction

Recent high profile cases – such as charities serving as conduits to finance terrorist activities and operations, insider deals with insufficiency transparency and charities engaging in abusive tax shelters – have drawn significant attention to non-profit transparency (Behn et al., 2010). The proliferation of unethical behaviors, insufficient safeguarding of recourses and criminal behaviors have resulted in a confidence crisis that outweighs the excellent work done by civic- minded groups. The importance of the public availability of audited financial information from non-profit organizations has been emphasized by the US Senate Finance Committee, because the non-accessibility of this information may result in a loss of public confidence (Behn et al., 2010).

The issues regarding accountability and responsiveness toward stakeholders and funders in non-profit organizations (NPOs) has also been highlighted in prior literature (Jang & Feiock, 2007). Transparency can be understood as the foundation of several accountability relationships.

Thus, a lack of transparency correlates strongly with a lack of accountability (Striebing, 2017).

Despite the prediction that the accounting methods used in the public finances will lead to more transparent annual reports, the organizational and financial transparency factor for NPOs are relatively underdeveloped compared to for-profit organizations (Harris & Neely, 2021a).

Because public trust is maintained through good governance and transparency (Harris & Neely, 2021a), a basic level of transparency disclosed publicly through financial records is encouraged by many countries such as the Netherlands and the US for NPOs.

The public continuously demonstrates an increasing amount of interest in publicly released information in the healthcare sector. Governments support the release of information to the public due to the assumption that if proper choices are to be made, and competition in the healthcare sector is to work, all parties participating in the healthcare market must be entitled to the same comparative information on cost and quality (Ossebaard et al., 2012). This information is necessary for consumers to explicitly compare the performances of healthcare providers and make an informed choice (Ossebaard et al., 2012). In 2001, the Dutch government launched the first Dutch National Portal with the expectation that an online information outlet on the

healthcare sector would increase transparency and encourage citizens to play a greater role in the development of a consumer-centered healthcare system (Ossebaard et al., 2012). This is an example of how the government encourage the NPO’s to be transparent. Therefore, the research question proposed for this study is as follows: Does the financial composition has an impact on

(6)

the transparency of NPO?

The present study therefore aims to evaluate whether the concept of NPO finances has an impact on accounting transparency. First, instead of focusing on how the funds are being spent, I examine the effect of funds on the level of transparency. To generate enough revenue to achieve their goals, NPOs rely mostly on donations and grants. Due to the donor reliance factor of these organizations, these organizations are expected to be more transparent, thus maintaining public trust. Secondly, I examine the effect of organization size on transparency. The aim is to evaluate whether the largest organizations are more capable of generating fund, and thus, these

organizations must be more transparent to demonstrate accountability to the stakeholders. The last construct studied in this paper is the effect of audit quality on transparency. Previous

literature positively associates an annual report that is audited by the Big Four auditors – such as KPMG, PwC, Deloitte and E&Y – with a high-quality audit. According to Beisland et al. (2015), high-quality auditing is particularly important for companies in need of financing. This paper aims to evaluate whether this factor also play a vital role in the transparency of NPOs.

In the past, several studies have focused on various principles in accountancy areas that reduce agency conflicts to mitigate managers’ opportunistic behavior and information

asymmetry among managers and outside shareholders (Makhlouf et al., 2018). García-Sánchez et al. (2017) argue that an effective board of directors mitigates agency conflicts and introduces greater transparency to financial reporting (García-Sánchez et al., 2017; Makhlouf et al., 2018).

They also note that for the purpose of monitoring managerial activities, board members must base their decisions on verifiable information; as such, boards of directors should insist on high- quality reporting. The one factor that all these previous studies share is that their main focus is mostly based on the transparency theory of for-profit organizations.

Previous literature in the non-profit sector has largely focused on how successfully these organizations raise funds and how efficiently these funds are spent. The main focus from other studies is on the distribution channels of NPOs. To contribute to the literature of metrics to evaluate firm performance, the main focus of this thesis is to investigate the determinants for the level of transparency in the third sector. The importance of having adequate organizational metrics to rely on, and to evaluates the organization performance level of transparency has been highlighted constantly. Thus, this thesis can contribute to a further step in providing the

stakeholders with a greater level of accountability and credibility from the non-profit sector.

(7)

Overall, the results of my study shows that medium to larger NPOs are better able to attract volunteers and make effective use of the resources available to generate contributions from donors. These contributions can be used to maintain public trust and enhance the credibility of the financial report, which is audited by the Big Four. The variables discussed in this study have a greater impact on the effort invested in transparency than the willingness to be

transparent. There are several limitations in the scope and methodology of this research study that should be taken into consideration due to the confidence level implemented to choose the random sample and the potential possibility of errors in my hand-collected data. The results of this study can only be interpreted with 90% assurance of the validity of the population.

(8)

2. Literature review

The first part of this chapter provides an overview of the current state of knowledge regarding agency theory, accounting transparency and non-profit organizations. The second part of this chapter presents and discusses the hypotheses of the study, which are formulated based on the theoretical framework presented in the chapter’s first section.

2.1 Agency theory

An agency relationship can be defined as a contract under which a person or a group of people perform certain services on behalf of another person (Jensen & Meckling, 1976). The person who delegates decisions to another person is called the principal, and the person authorized to act on the principal’s behalf is called the agent. According to Jensen and Meckling (1976), ensuring that the agent makes optimal decisions in accordance with the principal’s wishes is nearly impossible at zero cost.

The relationship between the stockholders and managers of a corporation is an example of what is understood with agency relationship and is mostly associated with the issues of

separating ownership and control (Jensen & Meckling, 1976). They further argue that the general problem associated with the use of an agent is what incentives the principal is going to use to motivate the agent to behave on the interest of the principal, therefore maximizing the welfare of the principal rather than his own. The previous literature has also focused heavily on the study of appropriate incentives to motivate the agent to make choices maximizing the principal welfare given the fact that uncertainty and imperfect monitoring exist (Jensen & Meckling, 1976). In the private sector, the accounting system has been positively related to the agency problem due to the fact that the primary function of accounting is to produce information for external users who have no access to internal records or special purpose reports (Pina et al., 2009).

2.2 Accounting transparency in corporate reporting

Accounting transparency can be understood as standard corporate communication with public stakeholders. Bushman et al. (2004) describe corporate reporting as a recurring report containing firm-specific information that the board of directors provides, either voluntarily or out of

(9)

obligation. The transparency of corporate reporting can be understood as a measure concerning a series of disclosures of general information included in a firm’s financial reports

Chiang and He (2010) state that from the company’s perspective, transparency helps mitigate agency problem and attract investments while lowering the cost of capital. For those outside the company, such as governments and corporate investors, transparency facilitates an understanding of company operations and governance (Chiang & He, 2010). Chiang and He (2010) also note that the board of directors is responsible for ensuring the efficiency of governance and supervision with the goal of enhancing company transparency, the effectiveness of governance and the quality of information.

The study from Bushman et al. (2004) considers the following four types of disclosures, which are measured by intensity to determine a firm’s level of accounting transparency:

- Governance disclosure is described by Bushman et al. (2004) as a disclosure provided in a report that explains the governance of the firm. The factors discussed include the affiliations and identities of the firm’s managers, board members and major shareholders, as well as information concerning employees’ and directors’ ownership of shares.

- Financial disclosure refers to the extent to which accounting methods, capital

expenditure, subsidiary information and research and development are disclosed in the report.

- Accounting principles refers to the type of accounting principal used in the report. Factors that should be considered are how general reserves are used and the extent to which subsidiaries on a consolidation basis with the firm are reflected in financial statements (Bushman et al., 2004).

- Credibility of the disclosure is an assessment of the reliability of a firm’s financial

disclosure. Bushman et al. (2004) observe that financial report audits conducted by one of the Big Four accounting firms tend to have high audit quality, which improves the

credibility of financial accounting disclosures.

Thus, the details discussed in these disclosures add to the level of transparency and accountability displayed in the annual report (Bushman et al., 2004). These disclosures are extremely useful to investors monitoring managerial decisions. Due to information asymmetry in the market, and to retain a solid relationship between the company and its investors, unbiased communication

(10)

protocol is necessary (Chiang & He, 2010). Thus, information transparency is a fundamental form for communicating the firm’s operating result to the stakeholders, and it is seen as a key reflection of proper governance (Chiang & He, 2010).

2.3 Non-profit organization

“Non-profit organization” is a term used to describe any type of legal entities that are not operated for profit and are not regarded under any particular legal system as part of the state sector (Bullain & Hadzi-Miceva, 2009). Some organizations categorized as NPOs include societies, associations, charities, foundations and non-profit corporations. The reasoning behind the existence of a non-profit organization can be divided into two categories: demand side or supply side theory. The legal and organizational environment and resources that encourage a non-profit organization’s foundation can be classified as supply side determinants (Luksetich, 2008). Social issues such as unemployment, poverty and population heterogeneity are classified as the demand side determinants of non-profit organizations. Overall, supply side determinants and government policies are more likely to encourage the existence of NPOs (Luksetich, 2008).

According to the author, there are three reasonings behind the fact that governments may subsidize NPOs rather than provide the service themselves. The first reason discussed by the author is the political advantage that governments may garner from the party. Political

advantages such as political support are a great encouragement (Luksetich, 2008). Secondly, the share cost of the operation may be lower for governments while NPOs are providing the service.

Finally, the volunteer labor available in the educational and healthcare sector results in lower costs during the provision of these services (Luksetich, 2008).

In the public sector, General-Purpose Financial Statements are used to satisfy financial and public accountability. According to Pina et al. (2009), to guarantee the affordability and sustainability of service delivery and the value of money within a city, government accounting aims to provide the citizens with crucial information regarding issues such as the city’s financial position, financial performance, service efforts and accomplishments. Furthermore, the authors argue that the principal-agent relationship in the public sector is more complicated than the principal-agent relationship in the private sector, as more layers of agencies separate the

principals in the public sector (citizens) from the agents (managers; Pina et al., 2009). Pina et al.

(2009) also claim that as a result of information asymmetry, bureaucrats have an information

(11)

advantage over the public, which may produce moral hazards and lead to adverse choices.

Luksetich (2008) argues that where asymmetry information is present, contract failures follow, and NPOs should provide a few assurances of superior quality service to the public.

2.4 Voluntary disclosure for non-profit organization

Harris and Neely (2021) distinguish between accountant transparency and voluntary disclosure.

According to them, accounting transparency relates to the access to some particular information in an effort to convey openness and accountability, while transparency encompasses the overall practice of providing users with multiple aspects of information regarding the organization (Harris & Neely, 2021). Thus, transparency as an overall practice can be seen as an NPO’s willingness to be transparent. Transparency is also perceived as a form of signaling based on the assumption that signaling is engaged to mitigate the information asymmetries that are

particularly high between non-profit agencies and donors (Harris & Neely, 2021). Alternatively, Harris and Neely (2021) describe voluntary disclosure as the pieces of extra information an organization chooses as an individual to disclose voluntary, thus creating more openness in the organization’s transparency as a whole. Thus, voluntary disclosure can be seen as an NPO’s extra effort to be transparent. To therefore measure and quantify the level of an organization’s extra input toward being transparent, the voluntary disclosure, the number of non-mandatory pages concerning the management decision and analysis in the annual report are being reviewed (Saxton et al., 2011). Irrespective to whether the information disclosed by the organization is required or voluntarily produced by the organization, organizations that choose to be transparent increase the visibility of their information (Harris & Neely, 2021).

According to Bullain and Hadzi-Miceva (2009), stakeholders consider corporate

governance and disclosure as fundamental measures to protect stakeholders. For the purpose of reporting and accountability to the stakeholder, NPOs must acquire at least a basic level of transparency, meaning the balance sheet and the statement of income and expenditure (Bullain &

Hadzi-Miceva, 2009). In this context the term “accountability” for NPOs is understood as an obligation or willingness by these organizations to accept responsibility for their actions and hold themselves accountable to their stakeholders (Bullain & Hadzi-Miceva, 2009). The authors further describe transparency as the willingness or obligation of NPOs to make data about their operations publicly available, including financial, programmatic and organizational data. A key

(12)

component of NPO accountability is demonstrating financial responsibility and mission-based performance (Saxton et al., 2011).Thus, if the accounts and records of the NPOs are readily open and available to public scrutiny by funders, beneficiaries and others, the NPO can be categorized as transparent (Bullain & Hadzi-Miceva, 2009). According to the authors, voluntary disclosure and finance-related information is a powerful tool for making such demonstrations.

2.5 Regulations in the Netherlands

According to the Dutch Civil Code article 285 book 2, a Dutch foundation is an organization that has neither members nor stakeholders (Koninkrijksrelaties). By law, Dutch foundations are not required to deposit annual accounts at the Dutch Trade Register unless the foundation is

maintaining an enterprise. Maintaining an enterprise refers to the commercial activities that must be registered in the Dutch Trade Register. If the foundation does not maintain a company, the only two documents the foundation is required to draw are the balance sheet and the statement of income and expenditure. Even though the foundation must produce these documents, it is not a legal requirement to deposit them publicly in the Dutch Trade Register. According to the Dutch Civil Code article 396 book 2, if the foundation is maintaining a company but only meets two of the three criteria mentioned in article 396, the foundation is still not required to publish its annual reports to the Dutch Trade Register (Koninkrijksrelaties). The three criteria of article 396 are as follows:

- The total of the balance sheet of the foundation should be less than €4.4 million.

- The net turnover of the company within the foundation should be less than €8.8 million.

- The number of employees at the company within the foundation should be less than 50.

Thus, it can be concluded that there is no clear answer to the Dutch foundation requirements regarding the drawing of their annual report, because the situation depends on the circumstances wherein the organization is operating, as mentioned in the criteria above. However, most NPOs that decide to draw their financial situation in an annual report published their annual report at least on their website. Thus, NPOs can still communicate with the public and show some accountability to them.

(13)

2.6 Hypotheses

Past research has shown that firm size is a significant determinant for disclosure levels in both the for-profit and non-profit sectors (Behn et al., 2010). Bhojraj et al. (2004) also argue that firm size is one of the most powerful variables in explaining disclosure. According to Desoky and Mousa (2021), one of the main reasons is that agency theory is positively related with firm size.

Thus, to ease agency conflicts, larger firms are more likely to voluntarily disclose more

information in the annual report (Desoky & Mousa, 2021). Desoky and Mousa (2021) also argue that larger companies disclose more information in their annual report with the anticipation that they may be better able to access financial markets (Desoky & Mousa, 2021). Instead of trying to access financial markets, NPOs rely mostly on grants from the government or private donors and revenue received from service programs. Thus, to reduce government intervention, larger NPOs are more likely to disclose more information. Previous research has further associated larger NPOs with better use of internet to disseminate information (del Mar Gálvez Rodríguez et al., 2012). The action of attracting and involving volunteers in NPOs is associated with the efficient use of available information (del Mar Gálvez Rodríguez et al., 2012). Moreover, del Mar Gálvez Rodríguez et al. (2012) argue that the budget – along with the organization’s volunteers and staff – influence the development of, for example, the NPO’s webpage, therefore influencing the online transparency of these organizations. Therefore, the first hypothesis of this paper considers if larger NPOs are more likely to disclose more information regarding their financial data, so they are able to generate more contributions.

H1: Larger NPOs are more transparent in their financial report than smaller NPOs.

Due to the results of fraud, more focus has been placed on the role of independent

auditors in monitoring corporations and rules such as the Sarbanes-Oxley Act of 2002 (Tate, 2007). According to the author, these rules are not directly implemented for NPOs, although these organizations encounter the same monitoring challenges as their for-profit counterparts.

She also states that NPOs have a large reliance on public support through donations; as such, it is crucial for these organizations to uphold adequate levels of monitoring (Tate, 2007).

Thus, to ensure the quality of financial reporting, external auditors are expected to play a monitoring role to reduce agency costs arising from managers’ opportunistic behaviors (Abid et

(14)

al., 2018). By doing so, the external auditors ensure the reliability of the financial statements.

However, NPOs do not have owners, and the audit choice is based on the contracting

environment, which is affected by a mix of their revenue sources and funds used (Tate, 2007).

The requirement for external monitoring changes proportionally as the contracting environment and agency cost change within an organization (Tate, 2007).

According to previous research, high-quality auditing is particularly important for companies in need of finances (Beisland et al., 2015). As the donors of NPOs obtain no direct gain from the charitable contribution they provided, and also have no control over how the funds are used, they rely more strongly on monitoring to assure their contributions were used

consistently with their goals (Tate, 2007). Thus, high-quality auditing is positively related to finance needs.

Many study have tested and shown that corporations and the users of financial statements believe that financial report audits conducted by one of the Big Four are positively related to high-quality audits (Tate, 2007). High-quality audits improve the credibility of a financial report.

The author argues that large audit firms have more resources to better train their employees, thus ensuring and providing higher quality audits. He also states that due to the larger client base, the larger audit firms are more independent from their clients (Tate, 2007). In this study, NPOs are expected choose Big Four audit firms to conduct their audits, thus assuring their donors that the quality of the monitoring is high and improving the credibility of their financial report. By doing so, they reassure donors that their contributions are used as intended and continue to raise funds.

Therefore, the second hypothesis of this paper considers whether organizations that choose Big Four firms are more transparent, ensuring the donors of high-quality monitoring and enhancing the credibility of the annual report to continue to attract support through contribution.

H2: Big Four auditors are positive related with the non-profit organization transparency.

The strong incentives to withhold bad news with the hope that subsequent events will turn in favor of the managers is a problem managers face continuously (Kothari et al., 2009). The authors further argue that some Chief Financial Officers delay unwelcome news disclosure with the hope that the unwelcome news may never need to be released if the firm’s status improves before the deadline to release the annual report. In the case of NPOs, the goal is not to maximizes

(15)

profit but to rather make small profits, avoid losses and maximize the recipients’ welfare (Balsam & Harris, 2018). Nevertheless, these organizations must maintain a financial backup plan to accumulate resources and acquire capital assets, smooth expenditures and provide funds for rainy days (Balsam & Harris, 2018).

A multitude of factors, including internal motivations and external influences, add to the complexity of decision-making regarding whether the government or donor wants to fund a non- profit organization (Harris & Neely, 2021).Therefore, the proportion of contribution received from the government or private donor to the total revenue plays an important factor in the amount of detailed information to be publicly displayed. Harris and Neely (2021) argue that if NPOs have a larger ratio of contributions to the total revenue, they are more likely to respond to a request of audited financial statements. Due to the fact that these organization are more reliant on donor support, the request to display additional information is more likely to be accepted by non-profits (Harris & Neely, 2021). These organizations are not operated by profits but have enough funds to provide superior quality service and help society is necessary. Thus, the relationship between financial performance and accounting transparency is study, testing whether the proportion between the contribution and the revenue generated from services performed has an impact on voluntary disclosure. It is expected that organizations performing well can generate enough funds considering that these organizations have already gained the public’s trust (Harris & Neely, 2021). Thus, in this study, a positive correlation between the ratio of contribution received and the voluntary disclosure to ensure accountability to funders is expected.

H3: Greater amount of contribution received from donor support have a positive impact on transparency.

(16)

3. Research methodology

The first part of this chapter provides an overview of the sample collected for this study, while the second part of this chapter discusses the transparency model introduced in this study and a brief description of the variable measurements used.

3.1 Sample

This study aims to determine whether the concept of NPO characteristics has an impact on the transparency level. To answer the research question posed for this study and test the proposed hypotheses, an archival research method was conducted. Many previous studies have indicated that transparency in NPOs varies due to the differences in state regulations (Ruijer & Meijer, 2016). Thus, results obtained in the US and Egypt may not be representative of Europe. The setting of this paper is the Dutch healthcare sector, and the sample includes foundations of assorted sizes, as well as foundations that are involved in different social matter, to prevent a specific factor within the sector from influencing the results in the study. Thus, the results can be generalized to all kinds of NPOs.

The method used to collect data for this study is based on random sampling and was hand-collected from the databases of the website company.info, including information about Dutch foundations that publish financial data in their annual reports. Random sampling distinguishes two types of sampling: probability sampling and non-probability sampling

(Acharya et al., 2013). In this study, non-probability sampling was used, such that the probability that a subject was elected is unknown to reduce selection bias in the study (Acharya et al., 2013).

The data collected for the sample is from 2019 due to the assumption that years are not an important determinant in this research. Furthermore, the annual report of 2020 has not yet been finalized and made publicly available. Thus, the 2019 annual report is the most recent data available.

For the duration of 2019, the database on company.info provides 701 unique observations for healthcare foundations. The population from which I drew my sample was reduced to

organizations that are appropriate in terms of data availability or standard measurements. I eliminated all the organizations that presented the received grants as a fraction of organizational expenses and kept the organizations that presented the received grants as a fraction of

organizational income. The grants received as a fraction of organizational expenses are not

(17)

categorized as part of organization revenue but as part of a loan. To accurately measure how much an organization relies on donor support, the revenue of the organizations must be considered instead of the loans. To determine the appropriate sample size to represent the population, a combination of precision, confidence levels, and variability are considered. The Yamane formula is used to estimate the sample size because it is applicable for a different population’s proportion levels and confidence levels (Adam, 2020). For this study, a confidence level of 90% was implemented to calculate the sample size, and the result shows that a sample size of 88 observations was needed. During the hand collection process, I chose to collect 100 observations instead of 88 to add to the comprehensiveness of the sample.

3.2 Variable measurement

To empirically measure the hypotheses studied in this paper, the transparency model documented by Harris and Neely (2018) was followed as the standard transparency equation model. I

adjusted Harris and Neely’s (2018) model by eliminating several variables that are not necessary for answering the research question proposed in this study. By studying the paper of Perego and Verbeeten (2015)^, I gathered the variables that may plausibly indirectly influence transparency and included them as control variables into Harris and Neely’s model to estimate the empirical model for this study. I removed some of the control variable introduced by Harris and Neely (2018) regarding US regulations and introduced the control variable used by Perego and Verbeeten (2015), which has the same setting as this study. The empirical model introduced in this paper is as follows:

(1) TRANS= b1 ORGSIZE + b2 BIGFOUR + b3 REVGENERATION + b4 BOARDSIZE + b5 AGE + b6 SUBSEC +a

(2) TRANSDUM= b1 ORGSIZE + b2 BIGFOUR + b3 REVGENERATION + b4 BOARDSIZE + b5 AGE + b6 SUBSEC +a

(3) TRANSTOTAL= b1 ORGSIZE + b2 BIGFOUR + b3 REVGENERATION +b4REVDUM+ b5 BOARDSIZE + b6 AGE + b7 SUBSEC +a

(18)

The dependent variables of this study are transparency disclosure (TRANS and TRANSDUM).

Transparency is defined as the overall practice to provide stakeholders and users with multiple aspects of information regarding the organization (Harris & Neely, 2021). The measurement discussed by Bushman et al. (2004) considers documents such as balance sheets, items from income statements and fund flow statements, as well as governance data and information on accounting standards. The details discussed in these documents add to the level of transparency and accountability displayed in the annual report (Bushman et al., 2004). According to the Dutch Civil Code, the only two documents Dutch foundations are required to provide are the balance sheet and the statement of income and expenditure (Koninkrijksrelaties). Thus, all the extra information discussed in the annual report in accordance with the documents discussed by Bushman et al. (2004) – excluding the balance sheet and the statement of income and

expenditure, which is required for NPOs – is seen as information that is voluntarily disclosed. To measure the TRANS variables, the number of total pages and the number of pages dedicated to disclosed non-mandatory information in the annual report were counted. This variable measures the extra effort the NPO made toward be transparent. The TRANSDUM is a dummy variable taking the value of 0 or 1. The value 0 indicates that the organization provide only mandatory information in its annual report, while the value 1 states that the organization provided non- mandatory information in its annual report. This variable measures the willingness of the organization to be transparent.

The independent variables introduced in this study are derived from the literature discussed by Harris and Neely (2021), and these variables are as follows: the size of the

organization, the finances ratio of the NPO and if the audit was conducted by a Big Four auditor.

To test whether NPO size impacts transparency level, the first independent variable introduced in the model 1 and 2 is the size of the NPO, ORGSIZE. In prior research, several proxies of firm size were used as measurements, such as turnover, total assets, number of shareholders and more.

According to Desoky and Mousa (2021), there are no criterion nor theory provided in the

disclosure literature to choose among the firm size proxies. However, del Mar Gálvez Rodríguez et al. (2012) state that the number of employees is one of the common measurements for

organizational size. In some NPOs, employees can also be seen as the number of volunteers. For this particular reason, to test my first hypothesis, the categories applied in the database of

company.info to indicate the number of employees and the size of the organizations are used as

(19)

measurements in this paper. Thus, ORGSIZE is an ordinal variable containing values between 1 and 12, thus determining organizational size. These ordinal variables values are applied in the database of company.info to indicate the number of employees in categories ranging from zero employees to more than a thousand employees. Table 1 provides more details about the

categories applied.

Table 1: Categories for the variable ORGSIZE 1 = Refers to organizations with zero employees.

2 = Refers to organizations with one employee.

3 = Refers to organizations with the number of employees from 2 to 4.

5 = Refers to organizations with the number of employees from 10 to 19 6 = Refers to organizations with the number of employees from 20 to 49.

12 = Refers to organizations with the number of employees from 1000 up to more.

Note: This table provide an overview of the ordinal’s variables applied in this study for the variable organizational size. These are values applied in the database of company.info. Retrieved from https://company.info/sbi/Q86

To test my second hypothesis, the second independent variable measures whether the auditor of the organization is a Big Four audit firm or not, BIGFOUR. Bushman et al. (2004) used the audit element as a measurement for the credibility of the financial disclosure. In prior research, Big Four auditors are valuable, as they signal high audit quality. As donors have no control over how the funds obtained from their contributions are used, they rely on reliable monitoring to assure the contribution were used consistent with the goal they were intended for. According to previous research, high-quality auditing is particularly important for companies in need of finances (Beisland et al., 2015). BIGFOUR is a dummy variable taking the value of 0 or 1. The

(20)

value 0 declares that the organization’s audit was not conducted by a Big Four audit firm, while the value 1 indicates that the audit was conducted by the one of the Big Four audits firms. Based on prior literature, I expect Big Four auditors to indicate enhanced credibility due to the high- quality audit assured in the financial accounting disclosures.

To test whether the proportion of contributions received from donor support have a positive impact on transparency, I introduced the independent variable REVGENERATION.

REVGENERATION represents the ratio between the contribution received by the government or a donor and the revenue generated of services performed. According to Balsam and Harris

(2018), performance in for-profit literature is primary based on market- and accounting-based measurements such as stock returns. These measurements are not applicable in the non-profit sector, and profits are not an appropriate performance measure for NPOs (Balsam & Harris, 2018). It is expected that organizations performing well can generate enough funds considering these organizations have already gained the public’s trust (Harris & Neely, 2021). Thus, to see how well the organization is performing, the ability to provide funds to guarantee superior quality service is used as measurement. The primary forms of revenue discussed in the literature are direct donations, governments grants and program revenue. In this paper, direct donations and governments grants are not distinguished due to the fact that both funds are collected from public, not generated from program service; they are referred to as contribution. Another reason for the decision to not distinguish between the forms of contribution received is because there is not sufficient data about private donors or donations in the collected sample to categorized it as an aspect for itself.

I included several control variables in the regression of this study; as introduced by Perego and Verbeeten (2015), they can plausibly influence transparency in Dutch NPOs that rely on donors. BOARDSIZE is one of the control variables introduced in this study. Dutch NPOs operate under a board structure known as a two-tier board structure. This structure indicates that the board of directors and supervisory board are separated by law (Perego & Verbeeten, 2015).

Thus, the structure of the board is not the main focus; however, board members play an important role in NPO accountability and may affect transparency disclosure (Perego &

Verbeeten, 2015). Various prior research has studied the effect of board characteristics on the levels of transparency; therefore, I dropped this variable and introduced it as control variable.

Thus, in this study, board size refers to both the board of directors and the supervisory board, and

(21)

it is measured by the number of members in the board. In prior research, the age of the

organization played an important role because older organizations obtain a track record and may have better financial or non-financial performance than younger organizations (Perego &

Verbeeten, 2015). Therefore, AGE is introduced as a control variable and measured by the number of years the foundation has been registered in the Dutch Trade Register. Lastly, the subsector the organizations are operating in may affect the governance of the organizations; thus, subsector fixed effects, SUBSEC, are also included as a control variable. The standard business classification code used in the database of company.info to categorize the hierarchical

classification of economic activities was used to identify the subsector the organizations are operating with a total of 27 subsectors.

TRANS = The measurement for transparency level;

= The number of pages dedicated on non-mandatory information disclosed in the annual report.

TRANSDUM = Dummy variable; 0 indicates only mandatory information and 1 indicates annual report contains non-mandatory information.

TRANSTOTAL = The number of total pages in the annual report.

b1 ORGSIZE = There are twelve categories indicating companies starting from 0 employees to more than thousand employees.

See table 1 for more details.

b2 BIGFOUR = Dummy variable; 0 indicates not audit by the Big Four and 1 indicates audit by the Big Four.

b3 REVGENERATION = Proportion of contribution on the total income of the organization.

REVDUM = Dummy variable; 0 indicates no contribution and 1 indicates contribution received.

b4 BOARDSIZE = Number of board members of the board of directors and the supervisory board together.

b5 AGE = Number of years the organization is registered by the Dutch Trade register.

(22)

b6 SUBSEC = The standard business classification code used in the database of company.info. See table 2 for more details.

Table 2: The standard business classification code for the sub-sectors.

8710 = Nursing home for intensive care or heavy medical treatment 8720 = Homes and day care centers for the mentally disabled

72193 = Research and development in the field of health and nutrition 86102 = General hospitals

86103 = Categorical hospitals

86104 = Mental health and addiction care with overnight stay

86221 = Practices of medical specialists and medical day treatment centers

86222 = Psychiatrist practices and day treatment centers for mental health and addiction care 86232 = Practices of dental specialists

86913 = Practices of psychotherapists and psychologists

86919 = Other paramedical practices (not physiotherapy and psychology) and alternative healers 86922 = Occupational health and safety guidance and reintegration

86923 = Preventive healthcare (no occupational health and safety guidance)

86924 = Medical laboratories, thrombosis services and other treatment-supporting research 86925 = Ambulance services and central posts

86929 = Collaborative organizations in the field of health care and other health care support services 87302 = Nursing home specialized for old people

87901 = Youth care with day care centers and overnight stays 88101 = Home care service

88103 = Support and guidance of the disabled 88991 = Outpatient youth care

88993 = Local welfare work

88999 = Other social advice, community centers and welfare cooperation 94993 = Support funds (not in the field of welfare)

94996 = Other NPO

94997 = Another advocacy cooperation

Note: This table provides an overview of the ordinal variables applied in this study for the subsector in which the NPO are operating in. These are values applied in the database of company.info. Retrieved from https://company.info/sbi/Q86

As mentioned earlier, the transparency model discussed by Harris and Neely (2021) was used to determine the effect of NPOs’ characteristics on the transparency of these organizations. To test the relationship between my variables, I used an OLS regression based on Model 1 when running tests for TRANS, and a logistic regression based on Model 2 when testing for TRANSDUM. The first hypothesis of this study assesses whether larger organizations are more likely to disclose more information than smaller organizations. I expected b1, the coefficient on the organization

(23)

size, ORGSIZE, to be positive, because the organization size is positively associated with the disclosure level (Desoky & Mousa, 2021).

The second hypothesis tests whether NPOs’ financial report audits performed by the Big Four auditors are more transparent than audits not performed by Big Four auditors. Therefore, if transparency is higher when the audit is conducted by one of the Big Four audit firms,

BIGFOUR, b2 is expected to be positive.

The third hypothesis tests whether the proportion between the contribution and total revenue generated from services performed has an impact on voluntary disclosure. It is expected that once an organization gains the trust of the public, the probability that these organizations will perform better increases considering that they can generate sufficient funding. Therefore, to find support for hypothesis 3 that organizations that rely on contributions disclose more

information voluntarily, REVGENERATION, b3 should be positive.

To provide robustness to the results of my study, I introduced an interaction term between the Big Four variable and the revenue generation variable in Regression Models 1 and 2, respectively. Instead of running two different regressions to measure the transparency levels of NPOs, I decided to estimate a robustness model in which the dependent variables transparency is measured by the total pages of the annual report, known as TRANSTOTAL in Model 3. To test in what subsamples the results of my second hypothesis hold, I introduced an additional dummy variable for REVGENERATION known as REVDUM.

(24)

4. Empirical results

The first part of this chapter discusses and presents the descriptive statistics and correlation results found in this study. The second part of Chapter 4 discusses the empirical results of the conducted study.

4.1 Descriptive statistics

Table 3 provides the descriptive statistics of this study, whereas Tables 4 and 5 provide the descriptive statistics regarding organizational size and the organization’s auditors in more detail.

As noted in Table 3, there is a probability of approximately 71% that an NPO provides additional information regarding their financial data in the annual report as an effort to convey openness toward the public. The results further show that the NPO voluntarily provides, on average, approximately 7 to 8 pages to increase the level of transparency it provides to the public. As seen in Table 4, the larger NPOs generally provide more additional information voluntarily than smaller NPOs. It also shows that medium sized NPOs with employees numbering between 200 and 499 are more likely to put greater effort into conveying openness.

Table 5 shows that 33% of the sample financial audit are conducted by the Big Four auditors. The results further show that there is a probability of 90% that audits conducted by Big Four auditors provide additional information regarding financial data, whereas only 61% of the audits not conducted by Big Four auditors provide additional information regarding their financial data. The annual report of the organizations whose their audits were not conducted by the Big Four provide, on average, five pages of additional information, whereas the annual reports of organizations with audits conducted by the Big Four average 12 pages of additional information. Lastly, from the results of Table 3, it can also be concluded that approximately 17%

of the contribution received by the donor support contribute to the level of additional transparency provided by the organization.

(25)

Table 3 Descriptive statistics transparency sample

Variable N Mean Std. Dev. Min Max

Trans 100 7.54 12.613 0 79

Transdum 100 .71 .456 0 1

Orgsize 100 5.44 3.616 1 12

Bigfour 100 . 33 . 473 0 1

Revgeneration 100 . 171 .331 0 1

Boardsize 100 5.28 2.598 1 14

Age 100 18.56 13.491 2 83

Subsec 100 77844 25796 8710 94997

Note: Refer to Page 22 for the definition of the variables.

Table 4: Descriptive statistics for the organizational size - Transparency

Orgsize Trans Transdum

0 employee

N 15 15

Mean 7.067 .488

Std. Dev. 8.379 .667

1 employee

N 3 3

Mean 2 .667 Std. Dev. 1.732 .577 2 to 4 employees

N 27 27 Mean 3.037 .593 Std. Dev. 3.345 .500 4 to 9 employees

N 5 5 Mean 7 .6 Std. Dev. 8 .548 20 to 49 employees

N 9 9 Mean 3.778 .778 Std. Dev. 3.073 .441

(26)

50 to 99 employees

N 1 1 Mean 43 1 Std. Dev. . . 750 to 999 employees

N 7 7 Mean 6.143 1 Std. Dev. 4.413 0 1000 up to employees

N 9 9 Mean 12.444 .889 Std. Dev. 16.272 .3333333 Total

N 100 100 Mean 7.54 .71 Std. Dev. 12.61346 .456048

Note: The descriptive statistics of this study are summarized by the variables trans and transdum. Trans is measured by the number of non-mandatory pages. The transdum variable is a dummy variable; 0 refers to only mandatory information, and 1 refers to contains non-mandatory information. The variables are categorized by the organizational dummy variables.

(27)

Table 5: Descriptive statistics for the non-profit organization’s auditors - Transparency

Big four Trans Transdum

Not audited by Big Four

N 67 67

Mean 5.150 .612

Std. Dev. 7.793 .491

Audited by Big Four

N 33 33

Mean 12.393 .909

Std. Dev. 18.185 .292

Total 100 100

Mean 7.54 .71

Std. Dev. 12.61 .456

Note: The descriptive statistics of this study are summarized by the variables trans and transdum. Trans is measured by the number of non-mandatory pages. The transdum variable is a dummy variable; 0 refers to only mandatory information, and 1 refers to contains non-mandatory information. The variables are categorized by the Big Four dummy variables.

In Table 6, the Spearman correlation matrix is illustrated. Interestingly, the first observation about Table 6 as a whole is that the highest correlation in the table has a value of (0.7455) and is between the TRANS variable and TRANSDUM variable. This correlation indicates that the willingness of the NPO to provide additional information voluntarily has a strong correlation with the effort and additional number of pages the organization provides in their annual report.

The rest of the variables show moderate to weak correlations toward each other. As noted in Table 6, the organizational size variable has a value of (0.2415) and (0.2136), which indicates a weak positive correlation with the transparency variables. The results suggest that

organizational size has a small influence on the willingness of the organization to provide information voluntary. The results are contrary to the expectation of my first hypothesis.

The Big Four variable also has a moderate to weak correlation with the transparency variables, presenting values of (0.2490) and (0.3079). These results suggest the number of years the NPO have been operating in combination with board size and the proportion of the contribution received have a moderate impact on choices the organization made toward the auditor decision. Interestingly, the subsector in which the NPOs are operating has a negative

(28)

correlation with auditor choice. These results also contradict the expectancy of my second hypothesis.

The revenue generation variable has a weak correlation with the willingness of the NPO to provide additional information voluntarily and has a value of (0.2557). However, to the willingness to provide additional information, the revenue generation variable has a moderate positive correlation with the numbers of pages the organization provide in addition to their annual report and has a value of (0.3592). These results suggest the number of years the NPO has been operating in combination with board size has a moderate impact on the capability of the NPO to attract donor support. The subsector the NPOs are operating in has a weak negative correlation with the proportion of contribution received. These results are also contrary to the expectations of my third hypothesis.

The correlation between transparency and all the other variables of my models has a value below 0.50, suggesting that the transparency model discussed in this paper provide information regarding various aspects. The results also indicates that the correlations are relatively weak to negligible, and no single variable explains a large proportion of the transparency in general nor the additional pages.

Table 6: Spearman correlation between independent and dependent variables

Trans Transdum Orgsize Bigfour Revgeneration Boardsize Age Subsec

Trans 1.0000

Transdum 0.7455 1.0000

Orgsize 0.2415 0.2136 1.0000

Bigfour 0.2490 0.3079 0.4342 1.0000

Revgeneration 0.3592 0.2557 0.2611 0.3309 1.0000

Boardsize 0.3141 0.3264 0.3302 0.4297 0.4609 1.0000

Age 0.3517 0.2178 0.2999 0.4137 0.3954 0.4637 1.0000

Subsec 0.1624 0.0826 0.0072 -0.1287 -0.0063 -0.0591 0.0022 1.0000

Note: Refer to Page 22 for the definition of the variables.

(29)

4.2 Regression analysis 4.2.1 Organizational size

To test my hypotheses, I performed one logistic regression to measure the willingness of the NPO to provide information voluntarily and an OLS regression to measure the extra input of the NPO to be transparent. The adjusted R-Squared of my two models are respectively (0.1473) and (0.261), which indicates that the second model discussed in this section has a stronger relation with the level of transparency than the first model.

My first hypothesis tests whether larger NPOs are more prone to disclose more

information regarding their financial data than smaller NPOs, with the expectancy that they are able to generate more contributions. As noted in Table 7, my organizational size variable was not significantly related to the willingness of the NPO to be transparent. This result contrasts what my first hypothesis predicted and is also inconsistent with the findings of Desoky and Mousa (2021). Therefore, the first hypothesis of my study is rejected; it could also be concluded that there is no evidence that shows that larger NPOs are more prone toward transparency in their annual report regarding their financial data than smaller NPOs.

Contrary to the results in Table 7, the organizational size variable is positive and

significantly related to the additional effort of the NPO to be transparent. This result aligns with the findings of Desoky and Mousa (2021) and the expectancy of my first hypothesis. Therefore, the first hypothesis of my study is conformed. It could also be concluded that there is enough evidence to show that larger NPOs are more likely to include additional information voluntarily in their annual report than smaller NPOs.

As mentioned earlier, there are no criteria nor theories provided in the disclosure

literature to choose among the organizational size proxies. The organizational size proxy used in this paper is different than the organizational size proxy used in the paper of Desoky and Mousa (2021). The proxy used by Desoky and Mousa (2021) was based on the total assets of the organization instead of the number of employees or volunteers. The action of attracting and involving volunteers in NPOs is associated with the efficient use of available information (del Mar Gálvez Rodríguez et al., 2012). Thus, I cannot discount that there may be a possibility that due to the size proxy difference, the results of Table 7 are not in accordance with the paper from Desoky and Mousa (2021).

(30)

Overall, these results indicate that NPOs’ numbers of volunteers or employees do not directly influence the decision of whether these organizations would like to be transparent or not.

Instead, the number of employees or volunteers an NPO could attract indicates the amount of recourses the organization has to be more transparent and provide, for example, more

information on their website or annual report. If a larger NPO decides to be transparent in their annual report, they have more recourses. Thus, larger NPOs are more capable of attracting more volunteers and have more knowledge to add to the quality of the annual report; therefore, they contribute to a more transparent annual report

Table 7: Logistic Regression results – Transparency based on provide information voluntarily

Expected sign Coefficient t-stat

Orgsize + 0.064 (0.69)

Bigfour + 1.146 (1.67)*

Revgeneration + 0.627 (0.74)

Boardsize + 0.202 (2.21)**

Age + 0.011 (0.53)

Constant -0.946 (-1.82)*

Subsec fixed effect YES

Observations 100

Adjusted R-squared 0.147

Note: Refer to Page 22 for the definition of the variables. Logistic regression was performed with the Transdum variable. *, **, *** indicate the significant level at 10%, 5% and 1%

confidence level, respectively.

Table 8: OLS Regression results – Transparency based on extra pages provided voluntarily.

Expected sign Coefficient t-stat

Orgsize + 1.012 (1.99)*

Bigfour + 8.998 (2.21)**

Revgeneration + 11.785 (1.68)*

Boardsize + -0.086 (-0.13)

Age + 0.091 (0.62)

Constant -4.170 (-1.14)

Observations 100

Adjusted R-squared 0.261

(31)

Note: Refer to page 22 for the definition of the variables. OLS regression was performed with the Trans variable. *, **, *** indicate the significant level at 10%, 5% and 1%

confidence level, respectively.

4.2.2 Big Four auditors

The second hypothesis considers whether the annual report of an NPO whose audit was

conducted by one of the Big Four auditors is positively related to the transparency of the NPO.

As noted in Tables 7 and 8, the Big Four variable is both positive and significant for the two aspects of transparency discussed in this study. Therefore, the second hypothesis discussed in this paper is confirmed, and in according with the findings of Tate (2007), Big Four auditors are positively related to being more transparent.

As seen in Tables 7 and 8, the coefficients for the number of additional pages are significantly higher than the coefficient based on the willingness to provide information voluntarily. The coefficients indicate that the NPO with an audit conducted by one of the Big Four auditors has a stronger association to the organization’s efforts to provide additional information than their willingness to be transparent. These results also mean that the annual report audits conducted by one of the Big Four auditors are generally more transparent and voluntarily provide more information regarding their financial situation. Thus, the Big Four auditors are a way for NPOs to communicate with their donors, assuring the donors that their funds are being used in accordance with the goals. As seen in Table 8, an annual report audit conducted by the Big Four is significantly more transparent than an annual report not conducted by them, as they provide more details in the annual report. This result also supports Tate's (2007) determination that the Big Four are positively related with high-quality audits that enhance the credibility of a financial report.

4.2.3 Revenue generation

Lastly, the third hypothesis tests whether the proportion of the contributions received by the donor support has a positive effect on the transparency of the NPO. As noted in Table 7, my revenue generation variable was not significantly related to the willingness of the NPO to be transparent. The results contradict the prediction of the third hypothesis as well as Harris and Neely's (2021) findings. Therefore, the third hypothesis of my study is rejected. It could also be

(32)

concluded that there is not enough evidence to show that the proportion of contributions received by donors was related to the willingness to be transparent.

However, the results in Table 8 show that the revenue generation variable is positive and significantly related to NPOs that decide to voluntarily provide additional pages in their annual report. This result aligns with the expectancy of the third hypothesis and the findings of Harris and Neely (2021) that NPOs that receive more contributions provide more additional information in their annual report. Therefore, the third hypothesis of my study is confirmed. It could also be concluded that there is enough evidence that shows that NPOs that rely on donor support are more likely to provide additional value-relevant information to ensure accountability to donors to continue generating enough funds.

Overall, it can be concluded that the results of Harris and Neely (2021) did not hold true, as NPOs relying on donor are not more willing to be transparent in a smaller sample. There is a possibility that if my sample was larger, the results may align Harris and Neely’s (2021) work. I cannot discount the possibility that different stakeholders have different effects on transparency.

The differences of results in these two models may also be due to the level of monitoring the donors want to see. As such, if the NPO decides to be more transparent, than the results seen in Table 8 apply. If the NPO is not willing to be transparent, then the extra effort is not

implemented.

The results in Table 7 show that the control variable, board size, is positive and

significantly related to the willingness to transparent in their annual report, whereas the constant of this model shows the subsector in which the NPO operates is negative and significantly related to the willingness to be transparent in annual reports. The result indicates that the board size of an NPO contributes to the decision to be transparent in the annual report, which aligns with prior literature stating that an effective board of directors helps mitigate agency conflicts by introducing greater transparency to financial reporting (Makhlouf et al., 2018). The subsector, however, does not contribute to the transparency factor and has almost no effect on this decision.

As noted in Table 7, my age variable was not significantly related to either specification of transparency introduced in this study. Contrary to Table 7, the results in Table 8 show that none of the control variables introduced in this model are significant to the decision to provide additional effort to be more transparent.

(33)

4.3 Robustness analysis

As seen in the section above, the independent variable Big Four is the only variable significant to both specifications of the transparency variable. The results in Table 9 show that the relationship between the auditor who conducted the audits of the annual report for the NPO and revenue generation is not significant. This result suggests that there is no effect between the auditor who conducted the annual report and the factor that the NPO received contributions from a donor nor the proportion of the amount of contributions received from revenue generated by the service performed. Therefore, the independent variable Big Four relates positively to the transparency of NPO’s annual report . It appears that amount of contributions received from a donor does not influence the auditor choice of the NPO. The constant shown in Table 7 is positive and significant, in accordance with the assumptions made by Tate (2007). The results support the findings of Beisland et al. (2015), demonstrating that high-quality auditing is particularly important for organization in need of finances. As such, the amount of contributions received does not influence the auditor choice but being transparent does support the donation decision.

Thus, to enhance the credibility of their annual report, an NPO may choose a Big Four to aid the donation decision. Therefore, the results of the robustness analysis performed in this study support the results of the transparency model discussed in section 4.2.2. It can be concluded that the robustness analysis also confirms the second hypothesis of this paper.

(34)

Table 9: Transparency OLS regression based on total pages

Coefficient t-stat

Orgsize 1.411 (1.76)*

Bigfour -4.479 (-0.34)

Revgeneration -4.635 (0.82)

revdum 7.552 -0.540

Boardsize -0.540 (-0.45)

Age 0.151 (0.72)

Bigfour to revenue dummy -4.231 (-0.28)

Bigfour to revenue generation 19.202 (1.01)

Constant 22.015 (2.04)**

Observations 100

Adjusted R-squared -0.006

Note: OLS regression was performed with the Transtotal variable with an interception term between Big Four variable and revgeneration. *, **, *** indicate the significant level at 10%, 5% and 1% confidence level, respectively.

(35)

5. Conclusion

This study evaluated what characteristics of an NPO play a significant role in the NPO’s

transparency. The main focus of the prior research regarding accountancy transparency has been on various principles that reduce agency conflicts to mitigate managers’ opportunistic behavior and information asymmetry among managers and outside shareholders in the private sector.

However, for the non-profit sector, the focus has been on how successfully these organizations raise funds and how efficiently these funds are spent. However, my study does not focus on how well these non-profit organization attract funds, but rather on the effect of the finance

composition on the transparency of an NPO. Therefore, the research question in this paper is as follows: Does the financial composition has an impact on the transparency of NPO?

First, I examined whether larger NPOs are more prone to disclose more information regarding their financial data than smaller NPOs, with the expectation that they are able to generate more contributions. I found that the organizational size of an NPO has no influence on their willingness to be transparent, whereas it does influence the effort to be more transparent.

Thus, medium to large NPOs are more capable of attracting volunteers or employee that can make efficient use of the information available than smaller NPOs. Therefore, when an NPO decides to provide more information, they have the resources to do so.

Next, I examined whether the annual report of an NPO with an audit conducted by one of the Big Four auditors is more transparent than an annual report not conducted by the Big Four. I find that the annual report audits conducted by Big Four auditors are generally more transparent and voluntarily provide more information regarding their financial situation. This finding implies that the Big Four auditors are used as signal by NPOs to communicate with the public and assure donors that the funds are being used in accordance with the intended goals.

Finally, I examined whether greater amounts of contributions received from donor support have a positive impact on transparency. I found that the composition of the revenue generated by an NPO has no impact on their willingness to be transparent, but it does influence the amount of effort to be transparent. This result implies that NPOs successful at attracting funds may spend more resources to improve the efficient use of information rather than the choice to be transparent.

Furthermore, the results of my robustness test shows that audits conducted by a Big Four auditor are positively associated with high-quality audits and, therefore, increase the

The effect of non-profit organizations’ characteristics on their transparency.

Full text