Amsterdam Business School
MSc Thesis
The relation between the Code of Good Governance and Earnings
Management in Higher Educational Institutions in the Netherlands:
the effect of the branch code good governance.
Name: Mouad Fetian Student number: 10459502 Date of version: 17th August 2015 Word count: 11714, 0
Supervisor: Ms. Elma van de Mortel Second reader: Mr. Sander van Triest
MSc Accountancy & Control, specialization Control
Statement of Originality
This document is written by student Mouad Fetian 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.
Abstract
In this thesis, earnings management in the higher educational institutions in the Netherlands is been researched. And thereby contributes to the existent literate on earnings management by expanding the setting from the profit sector to the non-profit sector. In this
thesis there is examined whether earnings management occurs in the higher educational institutions in the Netherlands and whether the code of good governance (2012) has significant effect on earnings management. The results are especially interesting for the
Ministry of Education, Culture and Science, because they control the universities and colleges in the Netherlands on financial indicators. In case of that educational institutions
adopt earnings management in order to manage their financial positions, the financial
statements wouldn’t be transparent and reliable. In order to research the presence of earnings management the model of Jones (1991) and
the model of Burgstahler and Dichev (1997) is used. For this thesis 316 financial statements from the time period 2008-2013 are downloaded from the website of the Ministry of
Education, Culture and Science. The results showed that higher educational institutions in the Netherlands managed their
financial performance downwards instead of upwards and the educational institutions use income minimization as earnings management strategy. The results showed no significant
effect of the code of good governance (2012) on earnings management.
Key words: earnings management, non-profit, higher educational institutions, Netherlands, code of good governance.
Table of content
Abstract ... 3
1.Introduction ... 5
1.1 Background ... 5
1.2 Research question ... 6
1.3 Motivation & Contribution ... 8
1.4 Structure of the thesis ... 8
2. Literature Review ... 9
2.1 Sector description... 9
2.1.1 Higher Educational Institutions in the Netherlands ... 9
2.1.2 Higher Education and Scientific Research Law (WHW) ... 11
2.1.3 The role of the Ministry of Education, Culture and Science (OCW) ... 11
2.2 Earnings management ... 13
2.2.1 Introduction earnings management ... 13
2.2.2 Earnings management in the non profit sector ... 13
2.3 Managerial incentives for earnings management ... 14
2.3.1 Agency theory ... 14
2.3.2 Resource dependency theory ... 16
2.3.3 Other incentives for earnings management ... 16
2.4 Earnings management techniques and strategies ... 18
2.4.1 Earnings management techniques ... 18
2.4.2 Earnings management strategies... 19
2.5 Code of governance ... 20 2.6 Hypothesis ... 21 2.6.1 Hypothesis 1 ... 21 2.6.2 Hypothesis 2 ... 22 2.6.3 Hypothesis 3 ... 22 3. Research ... 23 3.1 Dataset ... 23
3.2 Methods for detecting earnings management ... 24
3.2.1 The Models ... 24
3.3 Research design ... 27
4. Results ... 28
4.1 Output regression Jones Model ... 28
4.2 Results Hypothesis 1 ... 29
4.3 Results Hypothesis 2 ... 31
4.4 Results Hypothesis 3 ... 34
5.Conclusion, discussion and limitations. ... 36
5.1 Summary... 36
5.2 Conclusion, discussion and limitations ... 36
5.3 Contribution ... 38
References ... 39
Appendix 1: Summary of sample... 42
1.Introduction
1.1 Background
In recent years Higher educational institutions in the Netherlands have been subjected to more attention with respect to their financial and risks position. The Ministry of Education, Culture and Science (OCW1) inspects the financial and risk position of the Educational Institutions in the Netherlands, particularly because the OCW funds the educational
institutions. The most important goal for the educational institutions is to provide qualitative education and to serve as a public goal for the society. The educational institutions have to deliver qualitative education with a limited of financial resources, which can cause a lot of pressure on the educational institutions as to their financial as well as their non-financial performance. This pressure can lead to incentives for the management to manage their financial performance in order to keep the stakeholders, the Ministry of Education and the society satisfied and at safe distance. An example is a recent event in the Netherlands, where negative news about Amarantis college came to light. The Amarantis colleges were millions short to meeting their financial obligations. According to the investigation
commission, the management and supervisory board behaved inappropriately, and conflict of interest and self-enrichment where also suspected in this case (S.Lindhout, 2013).
Another recent event whereby the pressure on the educational institutions became clear is at the University of Amsterdam. Students and teachers protested against the management, because of the students and teachers lack of participation in the university’s decision-making. Due to financial pressure from the Ministry of Education the University of
Amsterdam has to cut certain study courses, where these studies are not profitable because of the low number of students. This causes a lot of pressure on the university, because on the one hand education and research is their main focus, while on the other hand their financial position is also important. This pressure can lead to certain decision making by the educational institutions whereby the financial performance of the institutions are managed in order to show an healthy financial position to stakeholders and in order to show no big profits or losses. Managing of financial performance is known in the literature as earnings management.
1.2 Research question
This thesis is about earnings management in Higher educational institutions in the Netherlands. Earnings management is described by Healy and Wahlen (1999, p. 368):as follows:
“judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the
company or to influence contractual outcomes that depend on reported accounting numbers”.
There is a large body of research about earnings management in the profit sector (Dechow et al, 1995); (Burgstahler en Dichev, 1997); (Healy, 1985); (Dehow and Sloan, 1991) in contrast, in the non-profit sector research on earnings management is limited. Current research about earnings management in the non-profit sector is mostly done in the hospitals segment (Leslie G. Eldenburg 2011); (Leone and Van Horn, 2005) and (Ballantine, Forker & Greenwood, 2007). Incentives in the profit sector are to increase profit or stock price, whereas a non-profit organization doesn’t have as main incentive to make large profits. The incentives for a non-profit sector would be to lower the cost of debt and manage the result towards a break-even point or even downwards instead of upwards (Verbruggen and Christiaens, 2012). Higher educational institutions are mainly funded by the government and another portion of revenue comes from tuition fees from the students. Higher
educational institutions are accountable for subsidies they receive from the government and have to justify the expenses to the public in their annual report. Higher educational
institutions in the Netherlands serve a public goal, and their main goal is education and research. The expenses are also in line to achieve this goal. Higher educational institutions have no incentives to generate large profits but no incentives to make big losses either. In case of a large profit the Ministry of Education will lower their funds because of the surplus and in case of a big loss due to high expenditures institutions will be held accountable for their (excessive) expenditures.
Earnings management is a way of influencing the financial performance by the management. Influencing the financial performance has a negative effect on the transparency and
reliability of the financial performance of an on organization. In order to come to a reliable and transparent financial statement of the higher educational institution, the code of good
governance was established in 2006. This code of good governance is a branch code of the corporate governance code established in 2003. The code contains principles of good governance and is equal to the law of Higher Education and Scientific Research Law (Wet op het Hoger Onderwijs en Wetenschappelijk Onderzoek - WHW). The code applies to all educational Institutions in the Netherlands. In 2012, the code was adapted according to advice of external commissions and the code is also adapted according to changes in the law of WHW. The code contains five chapters with codes for the executive committee, board of trustees, financial management and for social accountability. The following code 2.3.1 is from chapter executive committees, p.7:
“Any form or appearance of a conflict of interest between the university, including associated institutions, and members of the executive board will be avoided. Decisions to enter into translawions whereby a conflict of interest can occur require the prior approval of the board
of trustees”.
“The executive board is responsible for the quality and completeness of the published financial reports. The board of trustees will ensure that the executive board adequately fulfils
this responsibility”.
I would like to examine whether earnings management occurs within the Higher educational institutions and whether the code of good governance of 2012 has any effect on the
manipulation of financial performance. With this thesis I want to add to the existing the literature by researching earnings management in the non-profit sector and examine if the code of governance has had significant influence on managing financial performance by managers.
The research question for this thesis is as follows:
“Do Higher Educational Institutions in the Netherlands manage financial performance at all, and if so, is there any effect of the branch code of good governance on earnings
1.3 Motivation & Contribution
Research on earnings management in the non-profit sector is little exposed, especially in the educational sector, whereas, in the profit sector, countless researchers have examined earnings management (Dechow et al, 1995). There has been a small amount of research in the non-profit sector: authors like Leone and Van Horn and Leslie Eldenburg studied earnings management in the hospital sector. I would like to examine the literature by examining earnings management in the Higher Educational sector as a setting. I expect that in the non-profit sector, earnings management is also a tool for managing financial
performance. Non-profit organizations are accountable for their expenses, and there is no profit mechanism. I expect that Educational institutions in the Netherlands use financial performance management to lower their profit or reduce their losses by using real earnings
management.
My personal interest in this matter comes from my work experience: I work at the University of Amsterdam, and because the university is a public organization I wonder whether the managers use earnings management to manipulate the financial results. Also, I would like to examine whether the code of governance has a significant effect on the behavior of the managers
1.4 Structure of the thesis
This thesis contains out of the following sections; literature review, research method, results and conclusion. The thesis begins with an briefly outline of the sector description in order to describe the setting of this paper, followed with the literature review on earnings
management in section two. The third section consist of the research design, the research method is statistical and quantitative. In order to test the hypothesis there is used a regression of the Jones model (1991) and as well a test of the distribution of discretionary accruals by the model of Burgstahler and Dichev. The results of the research is described in section 4 followed with section 5 conclusion, discussion and limitations.
2. Literature Review
This thesis will focus on higher educational institutions in the Netherlands. I will start with a brief outline of the sector. After that, I will discuss literature on earnings management, incentives, techniques and the code of good governance.
2.1 Sector description
In the following section , a brief outline of the educational institutions in the Netherlands will be given, with some key facts. The briefly outline of sector description gives us more insight and information in the sector of the educational institution, which will be used later on in this thesis. Furthermore, a description of the law, and the role of the Dutch Ministry of education, culture and science.
2.1.1 Higher Educational Institutions in the Netherlands
The higher educational institutions in the Netherlands include a total of 18 universities and 37 colleges in 2014. The 18 universities have a total of 240.000 students and 47.000
employees, the colleges have 446.000 students and 40.000 employees (OCW, 2014). A large part of the grand total of expenses made by the educational institutions consists of
personnel costs, approximately 75% of the total expenditure. The remaining 25% consist of housing costs and miscellaneous costs. For a summary of all educational institutions see appendix 1.
The main focus of an educational institution is the investment in education and research and to provide qualitative education with their available resources. The institutions in the
Netherlands are mainly funded by the Dutch government, the government fund comprises approximately 70% of the total income of the educational institutions. The remaining 30% of income of the educational institutions, consists of 20% coming from tuition fees, with the remaining 10% being other incomes like sponsorship and services to local companies. The income are displayed in figure 1.1 and the expenses in figure 1.2.
The government funding is determined by the Dutch Ministry of Education, Culture and Science. The Ministry uses student numbers and study results to determine the government fund of an educational institution. [art. 2.6 WHW] The regulations and procedures is
Figure 1.1: Summary of income.
In this graphic the total income of the universities are represented, with distinction government funds,
miscellaneous funds, college tuition fees and income from third parties. The figures display the averages for the universities and colleges from 2006 until 2013.
Figure 1.2 Summary of expenses.
In this graphic the total expenses of the universities are represented, with distinction of personnel costs, depreciation costs, housing costs and miscellaneous costs. The figures display the averages for the universities and colleges from 2006 until 2013.
€ - € 50.000.000 € 100.000.000 € 150.000.000 € 200.000.000 € 250.000.000 € 300.000.000 € 350.000.000 € 400.000.000 HBO UNIV Government funds misc. Funds Tuition fees
Income from third parties Misc. Income € - € 50.000.000 € 100.000.000 € 150.000.000 € 200.000.000 € 250.000.000 € 300.000.000 € 350.000.000 € 400.000.000 HBO UNIV Personnel costs Depreciation costs Housing costs Misc. Costs Total costs
2.1.2 Higher Education and Scientific Research Law (WHW)
The law of higher education and scientific research contains all the regulations and
procedures for the educational institutions, students and the government. The law is very important, because it contains the regulations for the government funds, which is a large part of the educational institutions income . The law also contains regulations for the management and supervisory board: these are mentioned in chapter 9 for the universities and chapter 10 for the colleges. Chapters 9 and 10 contain principles and guidelines for the management and supervisory boards and the procedure for inauguration and composition of the board.
2.1.3 The role of the Ministry of Education, Culture and Science (OCW)
The Ministry of education, culture and science controls the educational institutions in order to comply with the law of higher education and scientific research law (WHW). The Ministry uses financial ratios to determine the financial health of educational institutions. Financial ratios play an important role in assessing the performance and financial position of the institutions. Although the financial ratios give no value judgment about the quality of
governance, the quality of education or the quality of the educational institutions as whole, it is nevertheless important that the expenses of the educational institutions are legitimate and appropriate. A healthy financial position is important in order to ensure financial
continuity of the educational institutions. Financial continuity is concerned with the question whether the educational institution is financially healthy in order to meet its financial
obligations in the short and long term, and using adequate planning and control. Legitimate expenditure of the funding is concerned with the question whether the educational
institution spent the money of the funds on those things permitted according to the law and whether the institutions are entitled to receive the money from the Ministry (OCW,2014). All the educational institutions have to submit their financial statement for the previous year to the Ministry on 1 July at the latest. Before submitting the financial statement to the Ministry, the educational institution’s accountant has to check the financial statement according to the specific demands of the Ministry. With the attachment of an audit report and an assurance report, the accountant sends the financial statement to the Dutch Ministry of education, culture and science. The Inspection of Education which is part of the Ministry
and carries the responsibility to checks the financial statements of the educational
institutions. The Inspection checks the financial statements on the basis of financial ratios , the most important key ratios used are liquidity, profitability and solvency (Inspectie van het Onderwijs, 2012). If the ratios are lower than predetermined minimum values, financial risk may be possible, and in such cases the Ministry will analyze both the financial statements and the annual report. If necessary, further investigation of additional information will take place. In case of financial risk, the Ministry will place the institutions under adjusted financial supervision. The educational institution must then make a financial recovery plan in order to solve the financial problems within a foreseeable time. In case the educational institutions cannot solve the financial problems the Ministry will intervene, and, in a worst case scenario, the management will be removed from their position and the institution will lose a large part of its funds.
Figure 2: Summary of key ratios.
This graphic illustrates the averages of the key ratios from 2006-2013 and the predetermined key ratios from the Inspection of Education.
The financial ratios can be an incentive for the management to manage the financial
performance in order to display an healthy financial position of the educational institution to the Ministry. With a healthy financial position the educational institution can rely on
government funding, which is the largest proportion of their income as seen in figure
0,00 0,50 1,00 1,50 2,00 2,50 3,00 Colleges Pre-determined colleges Universities Pre-determined universities Profitability Solvency Liquidity
1.1.There is a lot of pressure from the government and other stakeholders on the financial and non-financial performance of the higher educational institutions, which also can lead to earnings management in order to present an financially healthy organization.
2.2 Earnings management
In the following section, the existing literature on earnings management and, the techniques and incentives will be reviewed. I will use the term earnings management in this thesis because this is a well-known concept in the existing literature.
2.2.1 Introduction earnings management
In the literature there are different definitions of, and perspectives on, earnings management. Schipper (1989) defines earnings as:
“ Disclosure management in the sense of a purposeful intervention in the external financial reporting process, with the intent of obtaining some private gain (as opposed to say merely
facilitating the neutral operation of the process)” and Healy and Wahlen (1999, 368) define earnings management as:
“judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the
company or to influence contractual outcomes that depend on reported accounting numbers”.
One simple and short definition of earnings management is that of Mulford and Comiskey (2002): “The active manipulation of earnings towards a predetermined target”.
The definition of Healy and Wahlen (1999, 368) is widely used and frequently cited by researchers, and I will be used this as definition for earnings management in this thesis.
2.2.2 Earnings management in the non profit sector
Where as a large amount on research can be found on earnings management in the profit sector (Dechow et al, 1995); (Burgstahler en Dichev, 1997); (Healy, 1985); (Dehow and Sloan, 1991)), while the body of research on the non-profit sector is limited. Leslie G. Eldenburg et al. examines earnings management in the hospitals of California. They contribute to the
literature of earnings management by expanding the analysis to nonprofit organizations. The authors examine whether, and how, managers use real earnings management to manage earnings downward. They expect that managers have incentives to manage earnings downwards, because nonprofit organizations are tax exempt. The results of the study indicate that real earnings management occurs in the hospital sector in California. Similar research, by Leone and Van Horn examines accrual-based earning management in nonprofit hospitals. They hypothesize that nonprofit hospitals have incentives to manage earnings to a break-even point. The authors found evidence that nonprofit hospitals adjust discretionary spending to manage earnings and also discretionary accruals to manage earnings (Leone and Van Horn, 2005). Verbruggen and Christiaens researched whether earnings management occurs in the nonprofit sector and whether the nonprofit organizations are dependable from the government. The results showed that nonprofit organizations manage their financial performance downwards in order to maintain government funds (Verbruggen and Christiaens, 2012).
2.3 Managerial incentives for earnings management
In this section, the incentives for earnings management will be outlined, and earnings management techniques will be discussed. A large part of the literature is concerned with the question why earnings management occurs at all. Personal interest is an important and common incentive for earnings management in studies about incentives of earnings
management.
2.3.1 Agency theory
Earnings management can be interpreted as follows: managers use judgment in their reports to mislead stakeholders to lead them to believe in the financial position of the organization. This situation can occur due to information asymmetry and can lead to earnings
management (Healy and Wahlen 1999).Information asymmetry occurs when the management has more information than the stakeholders do, and uses this additional information for their own interests. In order to achieve their own interests managers will use judgments in their report to mislead the shareholders and so achieve their goal.
The standard approach used to study earnings management is the agency theory (Walker, 2013). Earnings management can be seen as a product of the agency theory, and there are conflicting interests between the reporting party and the party to whom is reported. Jensen and Meckling (1976) define the agency relationship as follows:
"A contract under which one or more persons (the principal (s)) engage another person (the agent) to perform some service On Their Behalf-which Involves Delegating some
decision-making authority to the agent."
Motivations for earnings management in the nonprofit sector cannot be transferred one on one from the profit sector to the non-profit sector. Because there are differences in
ownership structures between profit and non-profit organizations, there are also differences in the principal-agent problems. A study by Ben-Ner and Ting (2008) investigates the
differences in the agent-principal relationship between profit and non-profit organizations in the United States. The results show that there is another kind of agency problem which can differ according to the ownership structure of the organization. The agency problem would be greater among non-profit organizations. According to the researchers, this is because managers and their employees don’t have to make such an effort to achieve the goals of the organization within an non-profit organizations. Ben-Ner and Families (2008) argue that in profit-making organizations, there are more material incentives.
Higher educational institutions have, in contrast to profit organizations, a different and long, list of stakeholders: the national government, the public, teachers, students, the faculty, research partners, donors, local government and politicians. All stakeholders demand high quality education, with limited financial resources. Due to the pressure from different stakeholders, nonprofit organizations are more likely to adopt earnings management in order to achieve the desired break-even net income, instead of a large profit (Pilcher and van der Zahn, 2010).The interests of the stakeholders cause much pressure on the higher educational institution. In order to achieve a financially healthy position and a healthy financial ratio, mentioned in section 2.1.3, the management of a higher educational institution will make beneficial use of the information asymmetry in order to manage the financial results, so as to present a financially healthy organization.
2.3.2 Resource dependency theory
Another possible theory which can explain the incentives for non-profit organization to manage their financial performance, is the resource dependency theory (Helmig, Jegers and Lapsley, 2004).The theory describes how the organization is dependable on external
resources and how this dependability influences the behavior of the managers and the organization (Pfeffer and Salancik, 1978). The researchers studied the dependency of organizations, and the results showed a dependency on the main sponsor. In this thesis, the main sponsor of the colleges and the universities in the Netherlands is the Ministry of Education (OCW).
As mentioned in section 2.1 the higher educational institutions are highly dependent on the funding of the Ministry of Education. The Ministry also sets specific demands for the
expenditure of the funds, and expects that the funds will be invested in education and development. The educational institutions are also accountable for the expenditure of these funds and are obliged to report these expenditures. This is because either the funds might not cover all the expenditures, or the funds prove afterwards to have been too high, and either a surplus or deficit could occur.
It is also of great importance that educational institutions meet certain conditions and performance indicators to be eligible for these funds, which is an incentive for managers to manage their financial performance in order to break-even. In case of large profits, the government will lower the funding because of the surplus, and in case of a big loss due to high expenditures, institutions will be held accountable for their (excessive) expenditures.
2.3.3 Other incentives for earnings management
In the following section, the incentives for earnings in the nonprofit sector are described. The incentives for earnings management for the profit sector differ from those of the non-profit sector, but they are present nonetheless.
Statutory duty & tax regulation
The tax regulation in the Netherlands requires for nonprofit organizations to have either zero earnings or small profits in order to be eligible for a tax exemption. However, an nonprofit organization that shows structurally large profits will be considered to be a profit
organization, which can lead to the withdrawal of the tax exemption (Hofmann, 2007; Omer & Yetman, 2003, 2007). A large number of universities and colleges in the Netherlands are foundations and have the tax exempt statys. According to these regulations, the managers have strong incentives to manage their financial performance downwards in order to break-even so as to maintain this tax-exemption.
Debt covenant hypothesis
In order for an organization to obtain a favorable debt position, future income will be calculated as present income and expenses are postponed from the present to the future. This is described as the debt covenant hypothesis which indicates the importance for an organizations to present favorable financial ratios (Watts and Zimmerman, 1986). These hypothesis is also applicable for the educational institutions in the Netherlands. Because the Dutch inspection of education analyses the educational institutions by checking at their financial and risks position as mentioned in section 2.1.3. In order to meet the
predetermined financial ratios the educational institutions will adopt earnings management in order to reduce their cost of debt.
Politcal cost hypothesis
The political cost hypothesis emphasizes the influence of the politics on an organization. Political costs can be considered as withdrawal of the tax exemption, tax increases or curtailing the government funds (Watts and Zimmerman, 1986). This hypothesis applies to the educational institutions in the Netherlands, because educational institutions have a publicfunction and are highly dependent on the government funds. The politics are a very important stakeholder for the educational institutions: politics provide the funds for the institution and have a large influence on the studies and educational institutions. Politics or the so-called the Ministry of education, culture and science expects an efficient and
legitimate spending of the funds without compromising the financial condition and
continuity of the organization. In order to keep politics at a safe distance, the management would manage their performance to break even and in favor of the financial ratios.
Conclusion
From the literature review on earnings management incentives it is most likely that managing financial results is plausible in the higher educational institutions in the Netherlands. The Ministry of Education, Culture and Science examines the educational institutions on their financial performance, with the financial ratios as a control point. This causes a lot of pressure on the educational institutions, and this pressure can be linked to the political cost hypotheses and debt covenant hypotheses. The educational institutions are also highly dependent on funding by the Ministry which can be linked with the resource dependency theory. The information asymmetry gives the management the opportunity to manage financial performance to break-even with favorable financial ratios.
2.4 Earnings management techniques and strategies
This section discusses several earnings management techniques and strategies.
2.4.1 Earnings management techniques
According to Healy and Wahlen (1999), earnings management occurs when judgements by management take place. There are various ways managers use their own judgement in financial reporting (Healy and Wahlen, 1999):
Estimating of useful life and residual value
Estimating future pension benefits, deferred taxes, bad debt positions and asset impairments.
Making choices between LIFO or FIFO.
Inventory position, purchase management and accounts receivables policy. Making choices between making cost now or later in the future, for example R&D
cost.
Healy and Wahlen make a distinction between real earnings management and accrual-based earnings management. The first one, real earnings management, is the actual changes in income or expenses which have real impact on current or future cash flows; an example is by adjustment of production. The second, accrual-based, is the most common and cheapest way of managing the financial results. Healy (1985) defines accruals as: “I define accruals as the difference between reported earnings and cash flows from operations”. Another
definition by Bergstresser en Philipon (2006): “Accruals are components of earnings that are not reflected in current cash flows, and a great deal of managerial discretion goes into their construction”.
Accruals consist of discretionary and non-discretionary components. Non-discretionary accruals are the result of normal business activities and cannot be influenced, in contrast with discretionary accruals. Discretionary and non-discretionary accruals constitute the total of accruals. In this research, earnings management will be detected trough discretionary accruals, which is the most common, and several studies found evidence that results are managed by using discretionary accruals (Healy and Wahlen,1999; Walker,2013).
2.4.2 Earnings management strategies
In the literature about earnings management several strategies are described, one of the most cited strategies is the one by Scott (2006). The strategies of Scott (2006) can be described as follows:
Taking a bath
This variant occurs when an organization reports a larger loss than necessary. This often happens when there is a change in management or in case of reorganization; the losses are not caused by the actions of the current management. The intention is that the new board will start with a clean record and thereby increases the chance to show an improvement in the earnings in the next coming years. Taking a bath is an extreme form of income
minimization strategy.
Income maximization
This strategy income maximization is the opposite of income minimization and occurs when an organization wants to report maximum results, and is most likely to occur in the profit sector. Managers use this strategy because of the incentives to receive their bonuses and is in the profit sector a most common strategy.
Income minimization
In case of income minimization, the results are managed in such a way that the reporting year shows a low result. This situation occurs when an organization wants to send out the
wrong signal to the stakeholder, like politicians and stakeholders. This strategy can be linked with the political cost hypothesis (Watts and Zimmerman, 1986). This form could also apply to higher educational institutions whereby the results are managed to break even to send a signal to politics and to other stakeholders that all resources have been invested in
education and development. In case of a large surplus they would send a signal that the educational institutions can operate with fewer financial resources.
Income smoothing
Income smoothing is also known as income equalization and occurs when an organization does not want to show large results over long periods of time. This results in a saw tooth pattern where one year the results or higher and the next year lower.
Conclusion
It is more likely that In the nonprofit sector, management would manage the results downwards to break-even or just above zero (Pilcher and van der Zahn, 2010; Verbruggen and Christiaens, 2012). Thus the strategies of income minimization and income smoothing would best fit a higher educational institution. This expresses itself in the following
hypothesis; Higher educational institutions in the Netherlands are more likely do adopt income minimization or income smoothing as a strategy for earnings management.
2.5 Code of governance
The code of good governance is a code of conduct that applies to all Educational Institutions in the Netherlands. The code was initially obtained on 2 February 2006 for the Higher
educational institutions in the Netherlands and is based on the code of Commission
Takabsblat (2003). The code shows, in the form of principles and elaborate regulations and instructions to the executive boards and supervisory board, how to carry out their
responsibilities and tasks. The key point of the code is the separation between management and supervision. The association of colleges and universities supervises the universities and colleges on compliance with the governance code.
In 2010 the code was reviewed and adapted to the changes in the Higher Education and Scientific Research Law (Wet op het Hoger Onderwijs en Wetenschappelijk Onderzoek – WHW). In 2012, the new code was launched and it now contains new principles for the management and supervision as a guideline. Deviation from the code requires an explanation to the stakeholder and is known as the ‘apply or explain’ principle. The
guidelines of the code consist of transparency, responsibility and codes to prevent conflict of interest. The code also contains a section of social accountability.
The code contains regulations and principles for the management, among other things: transparency, social accountability and codes for good governance. The code should have a significant effect on the behavior of the managers and since earnings management is a way of influencing the financial performance and influences the transparency in a negative way, the behavior of the managers to manage the financial performance wouldn’t be according to governance code. The interesting question is to what extent the governance code influences the management and supervision board and whether the presence of earnings management declined after the code of governance was introduced.
2.6 Hypothesis
In previous sections the theory on earnings management was explained. In this section, the theory will be converted to a corresponding hypothesis.
2.6.1 Hypothesis 1
According to Healy and Wahlen, earnings management can occur due to information asymmetry and is the opposite of transparency and social responsibility. In case when earnings management occurs the code of good governance fails to fulfil its function. Earlier research in the non-profit sector showed that non-profit organizations manage their
financial performance to break even (Leslie G. Eldenburg 2011; Leone and Van Horn, 2005). The pressure from the Ministry of education, culture and science and other stakeholders may lead to incentives, influencing managers to adapt their earnings management techniques in order to obtain a healthy financial ratio and so maintain their funds. With these assumptions the first hypothesis is obtained:
H1 : The financial results of Higher Educational Institutions in the Netherlands are managed
downwards or upwards. 2.6.2 Hypothesis 2
According to Pilcher and van der Zahn it is more likely that in the nonprofit sector the management would manage the results to break-even. In the literature review of earnings management, different strategies for earnings management were described, namely the strategies of taking a bath, income minimization, income maximization and income smoothing. In case of a higher educational institution in the Netherlands income
minimization and income smoothing would fit the best as strategy for earnings management as mentioned in the conclusion of section 2.4. With the following hypothesis the question is whether the management of educational institutions use income minimization or income smoothing as strategy will be examined.
H2(1) : Higher Educational Institutions in the Netherlands use income smoothing as earnings
management strategy.
H2(2): Higher Educational Institutions in the Netherlands use income minimization as earnings
management strategy. 2.6.3 Hypothesis 3
The code of good governance is obtained for transparency of the financial reporting and encourages social accountability .In this part of the research, the central question is whether the code of good governance of 2012 has significant effect on the behavior of the
management and supervisory board and whether the management manage financial
performance for their own interest and against the principles and regulations of the code of governance.
In according to these assumptions the third hypothesis is obtained:
H3: Code of governance of 2012 has significant effect on earnings management in the Higher
3. Research
In the previous section, the theory on earnings management is described and the hypothesis are developed. In the following section, first of all the dataset is described then the methods on detecting earnings management and finally the research design is described.
3.1 Dataset
In order to research the hypothesis the dataset is obtained. The dataset contains of all the financial statements of all the higher educational institutions in the Netherlands. The
financial statements are downloaded from the website of the Ministry of Education, Culture and Science2 and then all the data is organized into an Excel file. These financial statements are published by the Ministry of Education, Culture and Science. For the process of
publication of the financial statements see section 2.1.3.
The data set consist of in total 64 higher educational institutions that operated in the time period from 2006 until 2013. The financial data from 2006-2013 is also all the available data on the website of the Ministry of education, culture and science. The data contains an Excel file with the balance sheet, income & expenses, cash flow and key figures of all the higher educational institutions. The years 2008 until 2013 will be used for this thesis, because of incomplete data in the years 2006 and 2007. In total 316 financial statements will be used as an observations in this thesis, each financial statement of a university or colleges of one year is an observation number. This implies there is no recognizing for mergers or for educational institutions that didn’t operate over the entire time period. The financial statement of 2008 will be used as zero value and therefore left out as an observation. The financial data of 2008 are necessary in order to calculate the dependent and independent variables for the
following years. In case an higher educational institutions didn’t operate in 2008 but in 2009 or forward, the first year will be used as zero value.
2 www.duo.nl
In following tables a summary of the dataset, for full details of all the observation including the year see appendix 1.
Figure 3. Summary of institutions
Institutions 2008 2009 2010 2011 2012 2013 Total
Colleges 38 38 38 36 36 36 222
Universities 13 13 17 17 17 17 94
Total 51 51 52 53 53 53 316
Figure 4. Summary of sample
Institutions
Financial-statements
Less first year N (incl. outliers)
Outliers N
Colleges 222 44 178 3 175
Universities 94 12 82 1 81
Total 316 56 260 4 256
3.2 Methods for detecting earnings management
There are different models which can measure earnings management, in this section the most commonly used models for the profit and nonprofit sector will be described and compared. To finally make a choice between the models that will be used in this thesis.
3.2.1 The Models
Model of Healy
The Model of Healy (1985) assumes that discretionary and non-discretionary accruals are inseparable, and the model calculates the total accruals. Healy (1985) recognizes the
distinction between discretionary and non-discretionary accruals, nevertheless, in the model of Healy there is no recognizing of the separability of discretionary and non-discretionary accruals. The model calculates the discretionary accruals by the difference in total accruals and non-discretionary accruals. The model estimates earnings management by deviations from normal levels of accruals by assuming that accruals are a constant fraction of assets.
The model of Healy (1985) is as follows : (1) NDA = TA / T
Wherein:
NDA: Non-discretionary accruals TA: Total accruals scaled by total assets T: year included in the estimation period Model of DeAngelo
Unlike the Healy model (1985) the basis of the model of DeAngelo (1986) is not the average total accruals but the total accruals of the previous year (t-1). According to DeAngelo the total accruals are unlikely to be constant over time due to nature of the accruals process and total accruals is only dependent on last year’s total accruals instead of the average of the years of estimation.
The model of DeAngelo (1986) is as follow : (2)
NDAt = TA t-1 Wherein:
NDA: Non-discretionary accruals
TA t-1: Total accruals previous end of year.
Model of Jones
The model of Jones (1991) is based on the model of DeAngelo (1986) and is the most
commonly used model. The model of Jones (1991) takes account of the economic conditions of an organization and is the first one to divide accruals into discretionary and
non-discretionary parts. The presence of non-discretionary accruals is seen as a sign of the existence of earnings management. A positive value is a sign of increasing earnings and a negative value is a sign of profit-reducing. Which make this model more applicable for this thesis to research whether the educational institutions manage their financial performance
downwards.
The model of Jones (1991) is as follow : (3)
TA /A t-1 = α + β1 ΔREV/A t-1 + β2 PPE t/A t-1 + ε
Wherein:
ε : Residual represents the discretionary accruals A t-1: Total assets end of year t-1
PPE t: Property, Plant and Equipment at the end of year t
Model of Modified Jones
In the article of Dechow et al (1995) "Detecting earnings management" in The Accounting Review an adjustment is made on the model of Jones (1991), with as result the Modified Jones Model. The model calculates earnings management by using an estimation of the discretionary component of total accruals. The adaption to the Jones model is made by the researchers because they consider the revenue to be also manageable by the managers through the accounts receivable balance. In the higher educational institutions the accounts receivable balance is the income of tuition fees and is not sensitive to manipulation. In this thesis the original Jones model would be best applicable.
The Modified Jones Model is as follow: (4)
NDAt = α1(1 / A t-1) + α2(∆REVt – ∆RECt) + α3(PPEt)+ vt Wherein:
vt : Residual represents the discretionary accruals A t-1: Total assets end of year t-1
PPE t: Property, Plant and Equipment at the end of year t ΔREV : Mutation in revenue between year t and t-1
ΔREC : Mutation in net receivables between year t and t-1
Model of Burgstahler and Dichev (1997)
The model of Burgstahler and Dichev is known as the distribution approach. This model focuses on a specific benchmark in case of this thesis the benchmark would be break-even income. The model tested whether the incomes are proportional distributed before and after mutation in discretionary accruals, thereby making it possible to compare the two distributions to investigate whether the earnings are manipulated.
The model is as follow: (5)
PRE-RESt = RESt – DA t Where:
PreRes t : Results before mutations of accruals RES t: Net results year t / A t-1
DA t : Total discretionary accruals in year t
3.3 Research design
In following section the research design for each hypothesis will be described. In previous section the models for this thesis are obtained. A regression of the Jones model (1991) will be used as well the distribution model of Burgstahler and Dichev (1997) in this thesis.
Hypothesis 1
To research hypothesis 1 is the Model of Jones (1991) will be used in order to test the obtained hypothesis in section 2.6. The model of Jones distinguishes between discretionary and non-discretionary accruals to determine earnings management. The discretionary accruals measure the degree of earnings management (Walker, 2013) In order to calculate the equation of the Model of Jones (3), first of all the total accruals will be calculated, with the following formula.
TAt = (ΔCurrent Assets - ΔCash –ΔCurrent Liabilities –Depreciation) / At-1 (6) After determining the total accruals, the components ΔREV , PPE and T/A t-1 of the Jones Model are calculated with the data of the financial statements. The residual of the Jones Model is the estimation of the discretionary accruals, as mentioned in the equation (3). By making a regression analyses in SPSS the residuals can be calculated. Whereby TA is the dependent variable and ΔREV and PPE the independent variables. In appendix 2 an
overview of the balance sheet items are described in order to calculate the dependent and independent variables.
Hypothesis 2
To research hypothesis 2 the model of Burgstahler and Dichev (1997) will be used, see equation (5) . This model compares the statistical distribution of the accruals, one before the mutation in discretionary accruals and one after the mutations in discretionary accruals. The model of Jones (1991) will be used in order to calculate the discretionary accruals. And as follow the discretionary accruals will be used in order to calculate PRE-RES and RES, which are the two statistical distribution’s. PRE-RES stand for the results before mutation in the discretionary accruals and RES is the net results after the mutation in the discretionary accruals. With the two statistical distribution it is possible to determine a pattern of strategy
for the educational institutions and in order to research whether the educational institutions adopt income minimization or income smoothing as earnings management strategy.
Hypothesis 3
In order to research the third hypothesis which is whether the code of governance has significant effect on earnings management, the independent t-test and the Mann Whitney U-test will be used. By using a dummy value “ 0” for all the observation before the
introduction of the governance code and dummy value “1” for all the observations after the introduction of the governance code in beginning of 2012. The independent t- test, tests whether there is a significant difference in the mean of the two groups 0 and 1. The Mann Whitney tests does the same and is for confirmation of the independent t-test.
4. Results
In section 3 the data and research design is outlined, and in the following section the results of the research will be described. First of all, I will discuss the results for hypothesis 1
according to the Jones Model (1991); after that the results for hypothesis 2 with the Burgstahler and Dichev (1997) model are outlined. Finally the results of hypothesis 3 are discussed according to the independent t-test and Mann Whitney U-test.
4.1 Output regression Jones Model
As mentioned in section 3.3 of the research design, the Jones model is used in order to determine the presence of earnings management. In order to determine the discretionary accruals, the total accruals are calculated with equation (6) and the dependent variables are calculated. The total accruals are the dependent variable and the ΔREV and PPE are the independent variables. With a regression analysis the estimation of the discretionary accruals are calculated. The following figure is the output of the regression model:
Figure 5: Summary output regression of Jones Model (1991). Coefficients Unstandardized Coefficients Standardized Coefficients 95,0% Confidence Interval for B Collinearity Statistics
B Std. Error Beta t Sig.
Lower Bound
Upper
Bound Tolerance VIF
-,410 ,017 -3,315 ,001 -,065 -,017
PPEt / At-1 -,550 ,019 -,182 -2,950 ,003 -,091 -,180 ,992 1,008
ΔREV / At-1 -,150 ,008 -,106 -1,726 ,086 -,031 ,002 ,992 1,008
Model Summary
R R square Adj. R square Sig. Durbin
Watson
,219 ,048 ,040 0,002 1,995
ANOVA Sum of
squares DF Mean square F
Regression ,033 2,000 6,359 0,002
Residual ,647 253,000
Total ,680 255,000
This graphic illustrates the output of the regression analyses with TA as dependent variable and REV and PPE as independent variables. In order to ensure a representative sample all output of DA greater than 3 and less than -3 are marked as outliers and removed from the sample size. And also to ensure a higher significance value and model correlation. For overview of outliers see appendix 1.
The summary of the output of the Jones model gives the total influence of the independent variables on the dependent variable. The model correlates 21.9% with the dependent variable and 4,8% of the total accruals can be explained from the independent variables. The low R square gives an indication that the results are not very reliable. The ANOVA indicates whether the model is significant and whether Jones model has added value . The significance value is 0.002, which indicates that Jones model has a certainty of more than 95% . The Durban Watson measures the autocorrelation and must be between 1 and 3, in order to have no autocorrelation. Autocorrelation measures the degree of similarity over a times series in the sample. In this sample a regression analysis is made over the entire data instead specifically for each educational institutions. Thereby it is important that there is no autocorrelation, the value is 1.995, so we can conclude that there is no autocorrelation in the sample.
4.2 Results Hypothesis 1
In order to test hypothesis 1 whether educational institutions in the Netherlands adopt earnings management in the time period 2009 and 2013, the output of the regression model is summarized in figure 6. To determine whether there is earnings management we
measure the degree of discretionary accruals. The discretionary accruals are calculated by a regression model of the Jones model (1991) and gives the proportion of earnings
management. In figure 6 it is visible that over the time period 2009 until 2013 the
educational institutions managed their results downwards or upwards. The discretionary accruals are not equal zero or around the point of zero.
According to these findings and assuming on a low reliable Jones model as mentioned in section 4.1, H0 can be rejected and H1 accepted.
H1 : The financial results of Higher Educational Institutions in the Netherlands are managed
downwards or upwards.
Figure 6: Output Jones Model in negative and positive DA.
Negative DA Positive DA 2009 Mean Std. Mean Std. Colleges -0,639210588 0,769375879 0,619891579 0,431673854 Universities -0,63902 0,783328934 0,467981667 0,593893998 2010 Colleges -0,842600556 0,599928952 0,714852778 0,705878345 Universities -0,557704444 0,464770811 0,444238 0,466276163 2011 Colleges 0,701196842 0,698794409 0,893317143 0,723982781 Universities -0,617338462 0,856569175 0,514534 0,256325185 2012 Colleges -0,420559167 0,36914476 0,926162174 0,899316508 Universities -1,174867273 0,874010164 0,771533333 0,492953272 2013 Colleges -0,739584167 0,441063781 0,637137391 0,520300116 Universities -1,0691225 1,283438745 0,757281 0,436733359
4.3 Results Hypothesis 2
In order to test hypothesis 2, the method devised by Burgstahler and Dichev (1997) is used, the formula for which was mentioned in section 3.2. With this model of statistical
distribution we can compare the discretionary accruals, one distribution before mutation in discretionary accruals and one distribution after mutations. In figure 7 and 8 the statistical distributions are shown. Figure 7 PRE-RES shows the results before mutations in the discretionary accruals and figure 8 shows the results after mutation in the discretionary accruals.
Figure 8: Histogram net-results 2008-2013
In the histogram of the PRE-RES the results are normally distributed and the most frequencies are around the point of the mean. Contrary to what is shown in the PRE-RES histogram, the RES histogram shows clearly that a large portion of the educational
institutions are around point zero after the mutation in the discretionary accruals. There is a large shift of the results to the break-even point in the histogram of results after mutation in discretionary accruals. A small portion of the educational institutions shows a small loss and small profit, the concentration is significantly high on the break-even point. As expected, the results of educational institutions are managed to break-even, which is supported by the distribution model of Burgstahler and Dichev (1997).
Figure 9: Descriptive statistics hypothesis 2:
RES PRE-RES N Valid 256 256 Missing 0 0 Mean ,436124 ,436123 Std. Error of Mean ,0346290 ,3830195 Std. Deviation ,5540700 1,0928310 Variance ,307 1,194 Skewness ,802 0,334 Std. Error of Skewness ,152 ,152 Kurtosis -0,807 0,287 Std. Error of Kurtosis ,303 ,303 Minimum -,3189 -2,5072 Maximum 2,1554 3,8768
Figure 9 displays the descriptive statistics. The means of both RES and PRE-RES are almost equal, instead of the standard deviation, whereby RES has a smaller standard deviation of 0.55 then PRE-RES 1.0928. This indicates a wider distribution of PRE-RES, which is
confirmed by the minimum and maximum of PRE-RES. RES is smaller, distributed with a minimum of -0.3189 and a maximum of 2.1554; this indicates that the results are approaching zero. RES has a skewness of 0.802 and PRE-RES a skewness of 0.334, which indicates that RES is more skewed to the left than PRE-RES. PRE-RES and RES are calculated using the model of Jones (1991)
According the model of Burgstahler and Dichev (1997) and the results of hypothesis 2, H2(1) can be rejected and H2(2) accepted.
H2(2): Higher Educational Institutions in the Netherlands use income minimization as earnings
management strategy.
There is evidence that higher educational institutions in the Netherlands adopt earnings management in order to manage financial performance. There is also evidence that the results are managed downwards instead of upwards. This conclusion is in line with the results of previous research by Verbruggen, S., Christiaens, J. (2012) and Leone Van Horn( 2005).
4.4 Results Hypothesis 3
In order to test hypothesis 3, the mean of the discretionary accruals before 2012 and from 2012 and forward will be compared and analyzed. All the financial statements before 2012 are group 0 and all the financial statements from 2012 and forward are group 1. Using an independent t-test and Mann Whitney U test the groups are compared in order research whether the code of good governance of 2012 has significant effect on earnings
management in higher educational institutions in the Netherlands.
Figure 10: Results independent t-test difference in mean DA
N Mean Std. Deviation Mean diff. T Sig.
2009 till 2012 151 -0,08450 0,9347
-0,2060 -1,633 0,104
2012 and 2013 105 0,1215 1,0711
Figure 11: Results Mann-Whitney U test
N Mean Rank
2009 till 2012 151 120,93
2012 and 2013 105 139,38
Mann-Whitney U test 6785
Z -1,961
Assymp. Sig (2-tailed) 0,05
The independent t-test examines whether there is a difference between the mean scores of the DA before and after 2012. A difference was found between the averages of the two groups, from 2009 till 2012 the mean discretionary accruals are negative. Which means that in this time period the educational institutions adopted discretionary accruals in order to manage the results downwards. This corresponds with graphic 7 of hypothesis 2. In 2012 and 2013 the mean of the discretionary accruals is positive and higher than the mean of group 0. The difference between group 0 and group 1 is not significant ( p=0.104>0.05). The Mann Whitney U test confirms the results of the independent t-test. The mean rank of group 0 (120.93) is lower than the mean rank of group1 (139.38). On the basis of these results it can only be assumed that there is a difference in the mean of discretionary accruals between group 0 and group 1, but there is no significant evidence that the code of good governance
influences these differences. Also, according to my expectations, the mean should go
towards zero in order to have no discretionary accruals. On the contrary however, from 2012 and onwards the higher educational institutions adopted more positive discretionary
accruals in order to manage their financial performance upwards. This was also visible in figure 7 and can have several causes: the year 2012 and onwards is characterized by an increase of students. This may have had significant influence on the organizations need to
show no loss.
To detect a trend reversal after 2012, there should be more years to research. This is perhaps something for further research. There is no significant evidence that the code of good governance has effect on earnings management, and so with this analyses and resulting information, H3 is rejected:
H3: Code of governance of 2012 has significant effect on earnings management in the Higher
5.Conclusion, discussion and limitations.
The following section provides the conclusion, discussion and limitations of this thesis. The research question will be answered from the research results and literature review. The results and literature review will also be discussed. Finally, the research limitations will be discussed and further research suggestions will be given.
5.1 Summary
The literature review shows that not only in the profit sector earnings management can occur, but also in the nonprofit sector (Leone and Van Horn, 2005; Eldenburg et al 2011). This thesis expands the literature by researching whether higher educational institutions in the Netherlands adopt earnings management, in order to manage their financial
performance. Because of the pressure from the Ministry of Education, Culture and Science, and the dependency on government funds, the expectations is that educational institutions manage their performance downwards instead of upwards. The resource dependency theory, political cost theory and agency theory support the motives for earnings
management in this setting. Besides the question whether educational institutions adopt earnings management, the second question is whether the code of good governance has a significant effect on earnings management. The code is there for transparency, responsibility and accountability, earnings management influences these components in a negative way. Thus the code of good governance can be seen as the opposite of earnings management.
5.2 Conclusion, discussion and limitations
This research was done in order to answer the following research question:
“Do Higher Educational Institutions in the Netherlands manage financial performance at all, and if so, is there any effect of the branch code of good governance on earnings
management?
In order to answer the research question of this thesis, the discretionary accruals were estimated according to the Model of Jones (1991). The model of Jones calculates the discretionary accruals which measure the degree of earnings management. With the model of Burgstahler and Dichev (1997) the distribution of the accruals before and after mutation in the discretionary accruals are compared. The output of the Jones Model is not reliable
enough with an R of 0,219 and an R square of 0,048. The model of Burgstahler and Dichev is based also an the output of the regression model, which makes the distribution model of Burgstahler and Dichev output also not very reliable. As in the paper of Leone and Van Horn (2005) and in this thesis a regression model and a distribution model is used. Unlike to the paper of Dechow and Sloan (1995) specific accruals are calculated. Specific accruals are defined by Healy and Wahlen (1991) and mentioned in section 2.4.1. Calculation of specific accruals is a more in-depth research and could give a more reliable output. Calculating specific accruals for one or more observation number in the dataset could give more insight in the earnings management techniques and a more reliable output. Examples of specific accruals are managing R&D costs and managing accounts receivable position. These data is available for further research with specific accruals instead of discretionary accruals. The models for detecting earnings management with discretionary accruals as mentioned in section 3.2 are most commonly used in the profit sector. A model specifically for a non-profit organization is not available in the literature, the models where initially obtained for the profit sector. Although the Jones Model (1991) recognizes the economic conditions of the organization and would fit the best for a nonprofit organization.
The conclusion drawn from hypothesis 1 is that higher educational institutions in the Netherlands manage their financial performance downwards or upwards. It must be mentioned that the output of the Jones Model is not reliable. For further research and a reliable output a regression model should be made for each educational institution and for each time period. Given the time frame of this thesis it is impossible to make a regression model for each or a sample of the educational institutions.
The model of Burgstahler and Dichev clearly reflects a shift towards zero of the results in the discretionary accruals before mutation and after the mutation, assuming on the not reliable output of the regression model. The strategy of income minimization is what is applied most by the higher educational institutions in the Netherlands: the results are managed towards break-even but not under zero. Which correspondents with the conclusion in the literature review in section 2.4. Pilcher and van der Zahn (2010) and Verbruggen and Christiaens (2012) hypothesis that in the nonprofit sector the management would manage the results downwards or just above zero. Reporting above zero is in line with the political cost
hypothesis in order to maintain government funds; after all, reporting loss would be a bad signal to give out to politics.
No significant evidence was found that the code of good governance has effect on earnings management from 2012 and onwards. It shows in the results of hypothesis 3 that the mean of the discretionary accruals in the period from 2012 till 2013 are higher than the mean of the discretionary accruals in the period from 2009 till 2012, this difference is not significant either. There any many factors that could cause the trend reversal in 2012. The output of hypothesis 3 is also based on the output of the regression model, which is not very reliable. For further research specific accruals can be calculated in order to determine whether the code of good governance has significance effect on the behavior of the managers.
5.3 Contribution
Research on earnings management in the non-profit sector is scarce, especially in the educational sector in the Netherlands. This thesis expands the literature by examining earnings management in the higher educational sector in the Netherlands and to my knowledge no research has been done about the effect of the code of good governance on earnings management in this setting. The results are especially interesting for the Ministry of Education, Culture and Science, because they judge the financial position of the educational institutions. The Ministry does not expect that the educational institutions manage their financial performance in order to maintain the government funds. Although the results are not significant the Ministry of Education, Culture and Science may want to take this in consideration and filter out such manipulation of financial performance. The association of universities and colleges might also be interested in whether the code of good governance has resulted in positive change for the educational institutions. According to the results there is no significant proof that managing financial performance declines after introducing the code of good governance. The association should maybe tighten the rules and give a specifics definition of earnings management to indicate the boundaries.
