What is the influence of the quality of governance on the financial
performance of Dutch hospitals?
Name: Pim van Deursen Student number: 10871950
Thesis supervisor: Elma van de Mortel Date: 15 February 2018
Word count: 12610
MSc Accountancy & Control, specialization Control
Statement of Originality
This document is written by student Pim van Deursen 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
This thesis is about the influence of governance on the financial performance of Dutch general hospitals. Prior research to quality of governance on financial performance is mainly done in for profit sectors. Not in the nonprofit sector. Prior research to this subject, has not often been done in the healthcare industry. The quality of governance is measured by 4 different quality variables. These variables are about the number of committees, the number of board meetings, the number of members on the board and the existence of board evaluations. The research is about 59 general hospitals in the Netherlands. Data has been collected from the year 2015. Some regression analysis have been made. Out of these analysis it appears that related to all hypothesis there was no significant result. However out of some additional analyses it appears to be that there are some significant results. The number of meetings of the supervisory board has a significant negative effect on the financial performance of small hospitals. The number of board members of the supervisory board has a significant positive effect on the solvency and profitability ratio. The existence of board evaluations of the supervisory board has a significant positive effect on the financial performance of small and medium hospitals.
Contents
1 Introduction ... 6
2 Healthcare industry ... 8
3 Theory ... 10
3.1 Performance measures... 10
3.1.1 Financial performance ratios ... 10
3.1.2 Solvency ratio ... 10
3.1.3 Current Ratio ... 11
3.1.4 Margin of profit ... 11
3.1.5 Debt service coverage ratio ... 11
3.3 Agency theory ... 11
3.4 Hypothesis ... 13
3.4.1 Hypothesis 1 number of independent supervisory board committees ... 14
3.4.2 Hypothesis 2 number of meetings per year of the supervisory board ... 14
3.4.3 Hypothesis 3 (number of members in the supervisory board) ... 15
3.4.4 Hypothesis 4 existence of board member evaluation ... 15
4 Research Methodology ... 16 4.1 Dataset ... 16 4.2 Dependent variables ... 16 4.3 Independent variables ... 18 4.4 Control Variables ... 19 4.5 Research method ... 19 5 Results... 21 5.1 Descriptive statistics ... 21
5.4 Regression analysis ... 26
5.4.1 Single regressions ... 26
5.4.2 Multiple Regression ... 27
5.5 Additional regression models ... 28
5.5.1 Different hospital sizes ... 28
5.5.2 Individual financial performance ... 29
6 Conclusion and discussion ... 33
6.1 Conclusion... 33
6.2 Discussion ... 34
6.3 Restrictions and further research ... 36
References & Literature list ... 37
Appendix 1 Box plots outliers ... 40
Appendix 2 List of hospitals ... 42
1 Introduction
In the last few years there were a lot of scandals in the Dutch healthcare industry. For example: Meavita, a very big healthcare corporation, went bankrupt in 2009. Meavita had over 20.000 employees and more than 100.000 clients. Meavita went bankrupt due to mismanagement of the top executives of the corporation (De Telegraaf, 2015). One of the members of the supervisory board reported no significant internal and external singnals to the other members of the board about the functioning of the CEO. In a letter to the House of Representatives in the Netherlands, Secretary Schippers presented some demands for quality of executives and supervisory board members for all health care corporations (De Volkskrant, 2013). She wanted to involve supervisory board members more in decisions of executives but only when these decisions have a big impact on the corporation. This means that the Secretary looks at the quality of governance.
This thesis is about the establishment of corporate governance. According to Merchant & Van der Stede (2007) corporate governance means that there is set of mechanisms and processes that helps to ensure that companies are directed and managed to create value for their owners, while fulfilling responsibilities to other stakeholders.
There have been some researchers which did research to the characteristics of the governance and the achievements of a corporation (Dey, 2008). Thompsen (2014) did research regarding the governance structure and non-profit organizations and find out that not very much research has been done. If we look at the structure of governance, we look at certain observable characteristics, such as whether a company has outside directors or the size of a supervisory board.
This thesis focusses on some governance quality variables and financial performance. Multiple researchers has done research to governance quality variables. For example: Boone et al. (2007) did research on board structure (size) and the impact on financial performance. An interesting and relevant paper to another phenomenon in the governance quality is the paper of Sherman et al. (1998). As Sherman et al. (1998) states, in order to better understand the role of the board, researchers should examine the board committees. Board committees are also important as a governance quality variable.
To get some other governance characteristics I will look at the paper written by Abbot et al. (2004) who investigated the impact of certain audit committee characteristics. These
characteristics are from the Blue Ribbon committee. The characteristics are aiming on improving the effectiveness of corporate audit committees. This will lead to better financial restatements. Some of these characteristics could also be important for the supervisory board in general.
Considering the fact that little research regarding the governance mechanisms of Dutch hospitals have been done before and the fact that this specific work field is under pressure after some financial scandals gives this thesis an added value. For example: Kallapur & Eldenburg (2003) did some research into hospitals in the United States. Of course there ben some economic papers related to Dutch hospitals but it is difficult to find research related to governance in Dutch hospitals.
After looking at all these characteristics and the motive for this thesis I came to the following research question: What is the influence of the quality of governance on the financial
performance of Dutch hospitals?
The rest of the thesis consists of six chapters. The next chapter discusses the healthcare industry. Next, the theory is discussed and after that the hypothesis is formulated in chapter 3. In chapter 4 the research methodology and data are described. Chapter 5 portraits the results. The conclusion and the discussion can be found in chapter 6 of the thesis.
2 Healthcare industry
As already mentioned the research is done in the healthcare industry of the Netherlands. Healthcare organizations are organizations which are involved in improvement of health via diagnosis, treatment and prevention of disease, illness, injury and physical and mental impairments. A healthcare organization is according to the law admissions care facility ( in Dutch: wet toelating zorginstellingen) an approved institution. A healthcare organization is a non-profit organization. That means it is not driven to make profit. Healthcare organizations are there for social purposes. The social purpose in this case is: taking care that healthcare organizations provide good care for people. In this thesis only hospitals in the Netherlands are included. The reason is that within the data of healthcare organizations there are also a lot of rehabilitation- and other care centers. These centers are all very divers and that makes it difficult to generalize them. The hospitals are a lot easier to generalize and most of the time they are bigger than other care facilities, so there is more data available.
The next part of the organization description is about the structure of hospitals. Hospitals are foundations or private companies with ministerial approval. Hospitals are called, as already mentioned before, approved institutions. Under current legislation hospitals are controlled independently. The Secretary gives herself the opportunity, by law, to intervene when a hospital is not adhere to legislation or norms and values. An hospital is an independently controlled entity and thereby it is easy to compare hospitals with regular profit firms. In particular in the sense of the board structure. The board structure within a hospital is related to the board structure of regular profit firms. In Dutch hospitals, there is a two tier board structure. The two tier board structure exists of two different kind of boards. The first board of the two tier structure is the “board of directors” which conduct the day to day management of a hospital. The second board is the “supervisory board” which conduct supervisory functions. The opposite of the two tier board structure is the one tier board structure. In the one tier board structure there is only one board which performs both management and supervisory functions. In Dutch hospitals there is a two tier board. In anticipation of the agency theory part of this paper the supervisory board is the principal and the board of directors are the agents.
Hospitals in the Netherlands could be divided into two kinds of hospitals. The General hospitals and the University hospitals. The general hospitals are only care hospitals but in the University hospitals also education is playing a role in the process. This thesis will be about the general hospitals because these hospitals are only focused on the healthcare process and not on the education process.
The general hospitals in the Netherlands are divided into three “size” groups by the firm of BDO(2016). Small, medium and big hospitals. The distribution of the hospitals is: 21 big hospitals, 18 medium hospitals and 20 small hospitals. The criteria for these groups are the beds in a certain hospital. The capacity is playing a role here. Small hospitals having a maximum of 350 beds. The medium hospitals having a bed capacity between 350 and 600 beds. The big hospitals have a capacity of 600 beds and more. The above criteria are from the investigation of BDO (2016). There are two forms of companies related to the general hospitals. First the foundation form, in the Netherlands are 55 general hospitals in the form of a foundation. The other 4 general hospitals are limited companies. These four hospitals are the following ones: Spijkenisse Medisch Centrum, Ommelander Ziekenhuis Groningen, IJsselmeerziekenhuizen and Slotervaart Ziekenhuis. This above knowledge is out of the database of the ministry of VWS. The differences in forms of companies are mainly differences in liability. Also tax regulations are different for foundations compared to limited companies.
There are some differences in hospitals in the Netherlands as stated above. The differences between the University and general hospitals. Also the forms of companies are different. However there are more foundations then there are limited companies. As already mentioned above in Dutch hospitals there is a two tier board structure. The differences in forms of companies does not change the governance structure. The different quality measures for the corporate governance that will be used in this thesis could thereby be used for limited companies and foundations. Relating the different forms to financial performance the BDO investigation uses the same ratios to compare the financial status of a Dutch hospital.
3 Theory
In the theory chapter of this paper I will explain what theory will be used in this thesis. In the first paragraph of this chapter there will be an explanation of the performance measures of a healthcare organization. In the second part the agency theory is clarified. The agency theory will be the theory on which this thesis is drawn upon. In the last paragraph of the theory chapter the hypothesis will be set out.
3.1 Performance measures
Performance measures are a part of a management control system. A management control system is, according to Chenhall (2003) a tool to provide information to support a manager by making his decisions. This information may relate to financial performance of an organization or the clients of an organization. Information could be about financial performance measures or about non-financial performance measures. In this paper it is about the non-financial performance of hospitals in the Netherlands. The paper provides information about financial performance hospitals taking the ratios of the BDO Benchmark Hospitals. This is an investigation of the accounting firm BDO to bankruptcy risks in the hospital section of the Dutch healthcare industry. They mentioned some financial ratios which, are related to financial performance of hospitals. In the following paragraph the ratios will be shown that are used to measure the overall financial performance of a hospital. Every financial ratio will also be shortly explained.
3.1.1 Financial performance ratios
In this thesis the financial performance ratios of the BDO investigation are used and explained. First the ratio is named, then the calculation and at last there is an explanation what the ratio is about.
3.1.2 Solvency ratio
Calculation: Solvency ratio = equity / total assets * 100%
The solvency ratio is about the extent to which an organization has buffers to absorb losses. It is about being capable of fulfilling financial obligations as a hospital. There are two types of solvency ratios. In the other solvency ratio it is about the debt part instead of the equity part as stated above. In line with the BDO investigation I chose solvency ratio 1. The norm for this ratio is 25% for the hospitals in the Netherlands (BDO,2016).
3.1.3 Current Ratio
Calculation: Current ratio = current assets / current liabilities
The current ratio is about the liquidity of the Dutch hospitals. It is about the extent to which the providers of the current liabilities could be paid from the current assets of the hospital. The norm for the current ratio is 1(BDO, 2016). The norm 1 means that the current liabilities could be paid out of the current assets. This norm is not only for hospitals but is also a well-known norm for normal profit firms.
3.1.4 Margin of profit
Calculation: Margin of profit = net result / total revenue
The margin of profit is the ratio which is about the relative profit margin of an hospital. The ratio should be around 1,4%, because this is the average of the investigation of BDO (2016).
3.1.5 Debt service coverage ratio
Calculation: Debt service coverage ratio = net operating income / Debt service The debt service coverage ratio, also known as the “debt coverage ratio” is the ratio of cash available for debt servicing to interest, principal and lease payments. The ratio of the debt service coverage ratio should be around the 2,35, because this is the average DSCR for Dutch hospitals. BDO uses 2,35 thereby as the norm.
Profitability
Calculation: Profitability ratio = net income / total assets
The profitability ratio reflects ratio of the net income and the total assets of a hospital. The profitability should be around the 1,5%, because this is the average of the profitability ratio for Dutch hospitals. BDO uses the 1,5% thereby as the norm.
3.3 Agency theory
In the paper of Van Puyvelde et al. (2016) there are two main theoretical perspectives explained to define the role of corporate governance. They define the agency theory and the stewardship theory. Van Puyvelde et al. (2016) briefly explains the agency theory as a situation where the owners (principals) and managers (agents) of an organization have different interests. Both principals and agents are more likely to act in their own interest. The stewardship theory, in contrast, assumes that managers and owners share interests. That is the opposite of the agency theory.
As already mentioned in the introduction part of this paper the supervisory board of directors of an healthcare organization were not handling in the best interest of the organization. Therefore, the agency theory fits more in this thesis. To deepen out the agency theory I want to define the agency theory on the basis of the paper of Jensen & Meckling (1976).
Jensen & Meckling (1976, p. 308) define the agency theory in the following way: “a contract whereby one or multiple people (the principal) an other person involves in order to carry out services in their interest (the interest of the principal) whereby decisions are delegated.”.
The main problem here is that the agent will not always handle in the best interest of the principal (Jensen & Meckling, 1976). According to Jensen & Meckling (1976) a principal can limit the deviation of the agent by making monitoring costs. Monitoring costs are for example having a board of directors expenses. Costs associated with issuing financial statements and employee stock options are also monitoring costs. The principal of an organization can pay the agent, just to be sure that the agent will handle in the best interest of the organization. Paying agents leads to, as so called, bonding costs. The agent will not always handle in the best interest of the organization and thus the principal, this leads to residual loss. According to Jensen & Meckling (1976) the relationship between shareholders and managers of an organization is a good example of a possible agency conflict. This could be a conflict because there is a separation between ownership and management. For example, in order to get more money for himself or more control in an organization a manager will do things which are not in the best interest of the firm and thus for the owners. However, a healthcare organization could also be a non-profit organization. When healthcare organizations are nonprofit companies they are mainly foundations. In that case there no real owner in the sense of the agency theory. Health care organization could also be in the form of a private company or a limited company. These forms of companies are more for making profit.
When a healthcare organization is a foundation, it is hard to distinguish an owner. Some papers are arguing with this statement. The paper of Du Bois et al. (2004) claims that there is ownership in a non-profit organization. The paper of Du Bois et al. (2004) investigated the right to monitor and control by the board of directors. Hence, it is typically assumed that the board can act as fiduciary of the donors, sponsors and the community as a whole and that it can therefore play the role of principal. We can say that the board of an non-profit organization can have philanthropic goals, but this does not mean that the managers of the organization are having the same goals. For a hospital this means that the supervisory board can act as fiduciary of the community and therefore can play the role as principal. This does not mean that the board of directors share the same interests. That is when the agency problem arise for a hospital. Or as
Steinberg and Gray (1993, p. 301) note “the absence of stockholders to whom profits are distributed does not mean that those who control a non-profit are free of economic incentives and motivations”. There are obviously agency problems within non-profit organizations. The board of directors of a hospital has also economic incentives and motivations to act in the best interest of themselves and not in the best interest of the hospital.
In the following part the possible agency problems within non-profit foundations (healthcare organizations) will be outlined. Also the possible solutions there are for these agency problems will be explained. For example, the Thompsen (2014) paper claims that, despite of the differences in organizational forms, non-profit organizations share many conventional agency problems with normal business companies. Thompsen (2014) named a few examples of agency problems within non-profit organizations. They mention that the principal in a non-profit organization is not the owner but the organization as a whole is the principal. Thompsen (2014 named the possible agency problem by paying excessive salaries to executives and directors. Thompsen (2014) also named the use of the organization’s assets on a more favorable basis then available to outsiders and that managers or board members are involved in competing organizations.
The description of the sector and the comments on the agency theory of various papers will be linked together. It is difficult to distinguish an owner for a hospital when this hospital is a foundation. As mentioned before a hospital is an approved institution by the Secretary of VWS, but this says nothing about any ownership of a Dutch hospital. An independently controlled institution, like a hospital, is thereby comparable with for profit firms. The board structure, as mentioned in the sector description part of this thesis, is comparable to normal firms. In the sector description part is stated that for a hospital in the Netherlands there is a two tier board structure. The two tier board structure is about that there are two independent boards within one hospital. The board of directors and the supervisory board. Regarding to the agency theory the board of directors is in this case the agent and the supervisory board is the principal.
3.4 Hypothesis
In this chaper of the thesis it is about the hypothesis. The hypothesis covers the quality of the governance and the financial performance of hospitals. The governance quality will be measured by 4 different variables. Some of them are related to theory, for example the theory of Dey (2008). Other variables are more field (or sector) specific. The performance will be measured by the different financial measures, who are mentioned before in section 2.2. For example the current ratio.
3.4.1 Hypothesis 1 number of independent supervisory board committees
The first governance variable is the number of different board committees in a supervisory board. This variable is also in the Dey (2008) paper. Dey (2008) named 4 committees which are likely to occur in a supervisory board. The audit committee, the compensation committee, the nominating committee and the governance committee. The audit committee is there to do financial checks and to control the financials of, in this case, a hospital. The committees exist of multiple board members. They focus on making members of the supervisory board responsible for different kinds of tasks. If these committees are absent as a whole in the supervisory board then these tasks are all performed by the entire supervisory board or not performed at all. Dey (2008) mentioned in his paper that when there are more committees, this should be better for the quality of governance. Christensen et al. (2010) mentioned that when there is a presence of an audit committee as well as a remuneration and a nomination committee this is positively related with accounting-based performance indicators. Relating this to the hypothesis, the hypothesis will be that when there are more committees this should have a positive effect on the financial performance of a hospital. H1 : A higher number of independent supervisory board committees has a positive effect on the financial performance
3.4.2 Hypothesis 2 number of meetings per year of the supervisory board
This governance variable is the frequency of board meetings, that is about the number of meetings of the supervisory board during one year. According to Dey (2008) the number of meetings during one year provides an indication of how effectively the board functions to the extent that this represents the regular attendance by the board to firm issues. In fact Dey (2008) says that it is about the effectiveness of the board. It is not said that if the supervisory board meets al lot, the quality of the governance will be better, because this could mean that the board is not efficient enough. Another paper that also finds something about board activity is the paper of Brick & Chidambaran (2010). Brick & Chidambaran (2010) find that there is no impact on the return on equity (ROE) when there is high board activity. Relating this to the hypothesis, the hypothesis will be that when are a higher number of meetings during one year this could have a negative influence on the financial performance of a hospital.
H2: A higher number of supervisory board meetings per year has a negative influence on the financial performance
3.4.3 Hypothesis 3 (number of members in the supervisory board)
The third governance variable is the size of the supervisory board. According to the papers of Lipton & Lorsch (1992), Jensen (1993) and Yermack (1996) the size of the board is very important in the quality of the governance. They found out that smaller board sizes are considered to be more effective in attaining higher monitoring. Smaller boards have less disagreements among board members, and are likely to be more efficient and organized in carrying out board functions, than larger boards (Yermack, 1996). Relating this to the hypothesis, the hypothesis will be that a greater number of supervisory board members has a negative influence on the financial performance of hospital.
H3: A greater number of members in the supervisory board has a negative influence on the financial performance
3.4.4 Hypothesis 4 existence of board member evaluation
The fourth governance variable is the evaluation of board members. For example the Kiel & Nichelson (2005) paper believes that board evaluations can help to prevent corporate failures. The paper of Manichilli et al. (2007) believes that board evaluations can contribute to effective boards and improved financial performance of firms. According to Dey (2008) firms undertake a lot of self-evaluations on an annual basis. Dey (2008) believes that the quality of the governance (board functioning) will be better when there is a form of evaluation. Relating this to the hypothesis, the hypothesis will be when board member evaluation exists it will have a positive influence on the financial performance.
H4: Existence of board member evaluations have a positive influence on the financial performance
These four hypothesis will be investigated in the next part of this thesis. Besides that, the research method will be explained. The dependent, independent and the control variables will be explained.
4 Research Methodology
In this chapter the research method of this thesis is explained.is. There will also be a description of the used resources. The independent, dependent and the control variables will also be clarified in this section.
4.1 Dataset
The sample exist of 63 hospitals in the Netherlands. There are 68 general hospitals in the Netherlands, however for 5 hospitals there was no information available regarding to the governance variables and were thereby not included in this specific sample. The governance variables can be found in the “DIGMV” Database of the ministry of VWS. In the database of the ministry all approved healthcare institutions of the Netherlands can be found. This is not only a database for information about hospitals but also about approved healthcare institutions in the Netherlands. It is pretty easy to filter out all general hospitals. I filtered this data regarding to the four specific governance variables and hand collected them and put the variables in my own sample. To collect the financial performance variables the BDO research regarding to the financial performance of hospitals in the Netherlands is used. This research was published in 2016 with data from 2015. For the remaining 63 hospitals the data was available. I hand collected all this data and put it in my own data sample. In this BDO research it is also clear what the difference in size is for Dutch hospitals. All data is from the year 2015. The reason that only data out of the year 2015 is used is because this is the most recent data available. To test the data of all 63 hospitals I used SPSS to get the descriptive statistics. The descriptive statistics will be in chapter 5 of this thesis. After testing the descriptive statistics there were some outliers which were influencing the skewness and kurtosis. The data is transformed after testing the descriptive statistics. In the end the sample exists of 59 general hospitals.
4.2 Dependent variables
In this paragraph the dependent variables of this investigation to governance quality and financial performance are explained. As already mentioned in the theory chapter of this thesis, the financial performance will be measured on the basis of 5 different financial ratios. These ratios are used by the investigation of BDO. BDO did research to the total financial performance of hospitals in 2015. This is about how hospitals are functioning in the financial field. In this investigation I want to link this to the quality of the governance of hospitals. In the table below I will outline the financial ratios, with their calculation and norm.
Table 1. Financial ratios, dependent variables
Ratio Calculation Norm
Solvency Equity / total assets * 100% 25%, according to BDO
Current Ratio Current Assets / Current Liabilities
1, according to BDO
Margin of profit Net result / total revenue * 100%
1,4%, according to BDO
Debt Service Coverage Ratio
Net operating income / Debt service
2,35 according to BDO
Profitability Net income / total assets * 100%
1,5%, according to BDO
The norms that BDO is using are mainly the average of the total hospitals per ratio. The ratios will be explained below. The formula to measure the total financial performance is the following one: Performance_financial = FIN_SOL + FIN_CR + FIN_MOP + FIN_DSCR + FIN_PRO
The solvency ratio should be 25%. This means that the equity should be at least 25% of the total assets of a hospital. The 25% is the norm that the firm of BDO is using.
The norm of the current ratio is of course 1, which means that a hospital is capable of paying their current liabilities with their current assets. If the current ratio is higher than 1, than the financial performance of a hospital is of course better.
The norm of the margin of profit is 1,4%. So when the ratio of a certain hospital is above the norm of 1,4% this hospital has a better financial performance.
According to the BDO investigation the norm of the debt service coverage ratio should be 2,35. When the norm is achieved by hospitals it means that there is enough cash available to pay for interest.
The norm of profitability should be around 1,5%. So the net income should be at least 1,5% of the total assets
To get a relevant number of financial performance, the number of all the financial variables is calculated into one relevant variable of all the ratios together. Which will be the total of all ratios. However, financial performance can also be measured by the ratios on their own. The total amount of the ratios does not show the whole picture of financial performance.
4.3 Independent variables
This research exists of two parts, the financial performance part and the quality of governance part. The financial performance part of this paper is already covered in the section of the dependent variables. The part of governance quality will be explained in this section. The governance variables are the independent variables of this investigation. In the table below the governance variables will be briefly explained and the way how they are measured is clarified. Table 2. Governance variables, independent variables
Governance variable Relevant SPSS name Relevant literature Number of independent
board committees
Board_Nr of committees Dey (2008)
Christensen (2010) Number of meetings during
one year of the supervisory board
Board_number of meetings Dey (2008)
Brick & Chidambaran (2010)
Number of member of the supervisory board
Board_number of members Dey(2008)
Lipton & Lorsch (1992) Jensen (1993)
Yermack (1996) Existence of Supervisory
board evaluations
Board_evaluations Kiel & Nichelson (2005) Manichilli et al. (2007)
4.4 Control Variables
The control variable that will be used in this thesis is size. Dey (2008) did research on the connection between governance structure and performance indicators. He also used size as a control variable. In the paper of Eisenberg, Sundgren and Wells (1998) they mentioned that the size of a company can influence the financial results of a company. For the hospitals in the Netherlands BDO looked at the size of the different hospitals. They compared the balance sheet of the hospitals and according to that they made a statement of what is a big or a small hospital. In the end they have recognized three different sizes of hospitals. There are big hospitals, medium hospitals and small hospitals. In the dataset they are labelled with the numbers one (Big), two(medium) and three(small). These sizes are determined by the amount of beds in a certain hospital.
4.5 Research method
The research method of this thesis will be the quantitative research method in the healthcare sector of the Netherlands. The research focuses on, as already mentioned before, hospitals. In this thesis the general hospitals in the Netherlands were chosen. The University hospitals in the Netherlands are not included in this thesis. These hospitals were not included because these kinds of hospitals have a University section which makes it hard to compare these hospitals with the general hospitals. For this investigation I will use the database of the ministry of VWS. This database is for the governance variables and were hand collected out of this database. The other data, also hand collected from the BDO investigation, is about the measuring of the financial performance. All the numbers that are included in this database are from the year 2015. Only the hospitals where all data is available are taken into the total population of this sample. To test whether or not there will be an influence of the governance variables related to the financial performance I will run a few regression analysis. This regression calculation will be written out below:
(Single regressions)
1) Performance_financial = a + βboard_nr of committees + βsize +e 2) Performance_financial = a + βboard_nr of meetings + βsize +e 3) Performance_financial = a + βboard_nr of members + βsize +e 4) Performance_financial = a + βboard_evaluations + βsize +e (Multiple regression)
Performance_financial = a + β1board_nr of committees + βboard_evaluations + βboard_nr of meetings + βboard_nr of members + βsize +e
First I will do the single regressions of the 4 single governance variables, to search for significant influence between those variables. Next I will do the multiple regression to see what the relationship between all the governance variables related to financial performance is. There will be also some additional regressions. These additional regressions are necessary because when the primary regressions are showing no significant influence it could be that these additional regressions do show significant influence.
The first additional regression is about the different sizes of hospitals. These size are investigated indepently from each other. The test is about the influence of the governance variables, of for example a small hospital, on the financial performance. The regression is shown below:
Performance_financial = a + β1board_nr of committees + βboard_evaluations + βboard_nr of meetings + βboard_nr of members + β(different)sizes +e
The second additional regression searches for influence of the governance variables on individual financial performance ratios. The regression is shown below:
Performance_financial(individual ratios) = a + β1board_nr of committees + βboard_evaluations + βboard_nr of meetings + βboard_nr of members + βsize +e
The last additional regression searches for influence of the governance variables on individual financial performance ratios, but the different sizes are investigated also seperately.
Performance_financial(individual ratios) = a + β1board_nr of committees + βboard_evaluations + βboard_nr of meetings + βboard_nr of members + β(different)sizes +e
5 Results
In this chapter the results of the research will be presented. First of all the data will be validated. The validation consists of the tests of skewness and kurtosis. After validation of the data the correlation test will be executed using the Pearson and Spearman correlation coefficients. At last the regression analysis will be executed to test the hypothesis and the results will be described.
5.1 Descriptive statistics
In the table below the descriptive statistics of all 63 general hospitals are shown.
Table 3 Descriptive statistics
SIZE Performance financial Board_nr of committees Board_nr of meetings Board_ evaluations Board_nr of members N Valid 63 63 63 63 63 63 Missing 0 0 0 0 0 0 Mean 2,00 3,8056 2,81 7,79 ,83 6,17 Std. Deviation ,842 2,40984 ,820 2,477 ,383 1,397 Minimum 1 ,69 0 2 0 2 Maximum 3 18,56 4 18 1 9
The financial performance exist of the total of all the financial ratios. The mean of all financial ratios is 3,80. The average of the number of board committees is 2,8. The average number of board meetings is 7,79. The means of the board evaluations is 0,83, which means that most of the hospitals in the Netherland have evaluations of their boards. The average board members on the board is 6,17.
The financial performance variable shows a maximum which is very high regarding to the mean value. This could potentially lead to skewness. This will be tested in the next chapter. In the following section the data is validated. The skewness and kurtosis are shown.
5.2 Skewness, Kurtosis and Outliers
At first, the research looks at the outliers in the data. With a box plot it is easy to investigate the outliers in the data. In the appendix of the is research there will be some box plots of the various variables of the data. If necessary, some box plots will be in the main text of this thesis. For the variable of existence of evaluations on the supervisory board it is not necessary to make a box plot because the outcome is only one or zero. The criterion for skewness and kurtosis is between -3 and +3. The criterion that is used is from the paper of Cain et al.(2017).
Graph 1. Box plot financial performance
For the financial performance there are a few cases in which there are outliers, namely case 19, 22, 62 ,63 upwards and 20 downwards. The outliers are showed in the box plot above.
Regarding the number of meetings of the board there are more outliers. 2, 21, 27, 32, 45, 52 upwards and 11 downwards. Regarding to the other variables there are no outliers worth mentioning.
When the above outliers are in the dataset there are some variables where there is a lot of skewness or kurtosis. You can see this in the statistics table below.
Table 4 Skewness & Kurtosis
Statistics Performance_ Financial Board_nr of committees Board_ evaluations Board_nr of members Board_nr of meetings SIZE N Valid 63 63 63 63 63 63 Missing 0 0 0 0 0 0 Skewness *4,296 -1,622 -1,756 -,285 1,192 ,000 Std. Error of Skewness ,302 ,302 ,302 ,302 ,302 ,302 Kurtosis *23,527 *3,675 1,119 ,546 *5,364 -1,599 Std. Error of Kurtosis ,595 ,595 ,595 ,595 ,595 ,595
* Skewness or kurtosis above +3 or under -3
In the table above states that the kurtosis as well as the skewness of the financial performance are very high. However, for the number of meetings we see that only the kurtosis is a bit too high. For a good research and testing of all the hypothesis, the data should be transformed to get a better skewness and kurtosis. When the data needs to be transformed, the variables with a lot of outliers should be transformed. In this case the variables financial performance and the number of board meetings. To get a better skewness and kurtosis, the decision to clear the following cases in the dataset has been made: case 12, 19, 62 and 63. These cases are chosen because these represent the highest outliers in the dataset. After this transformation there is a whole other skewness and kurtosis. This is shown in the statistics table below.
Table 5 Skewness & Kurtosis after transformation
Statistics Performance _financial Board_nr of committees Board_ evaluations Board_nr of members Board_nr of meetings SIZE N Valid 59 59 59 59 59 59 Missing 0 0 0 0 0 0 Skewness ,611 -1,534 -1,984 -,390 ,839 ,033 Std. Error of Skewness ,311 ,311 ,311 ,311 ,311 ,311 Kurtosis 2,358 *3,213 2,001 ,699 2,050 -1,591 Std. Error of Kurtosis ,613 ,613 ,613 ,613 ,613 ,613
After the transformation of the data, there is no more skewness in the dataset. There is just a little bit kurtosis. The kurtosis is in the variable of the number of board committees. It is just above 3. It is not worth it to transform the dataset again. The dataset is now normally distributed after the transformation.
5.3 Multicollinearity
To judge if there are no high mutual correlations between the variables this will be tested by the Pearson and Spearman test. I took all the variables in the same analysis to check if there was any significant correlation between certain variables. In the tables below you can find the Pearson and Spearman tests, the conclusion of these tests will be explained below these tables.
Table 6 Pearson correlation
Correlations SIZE Performan ce_financi al Board_nr of commitees Board_eva luations Board_nr of meetings Board_nr of members SIZE Pearson Correlation 1 ,030 -,271* -,178 -,043 -,344** Sig. (2-tailed) ,823 ,038 ,177 ,744 ,008 N 59 59 59 59 59 59 Performance _financial Pearson Correlation ,030 1 ,082 ,185 ,163 ,171 Sig. (2-tailed) ,823 ,534 ,160 ,218 ,196 N 59 59 59 59 59 59 Board_nr of commitees Pearson Correlation -,271* ,082 1 ,066 ,099 ,155 Sig. (2-tailed) ,038 ,534 ,621 ,457 ,240 N 59 59 59 59 59 59 Board_evaluation s Pearson Correlation -,178 ,185 ,066 1 ,185 ,142 Sig. (2-tailed) ,177 ,160 ,621 ,162 ,285 N 59 59 59 59 59 59 Board_nr of meetings Pearson Correlation -,043 ,163 ,099 ,185 1 ,064 Sig. (2-tailed) ,744 ,218 ,457 ,162 ,630 N 59 59 59 59 59 59 Board_nr of members Pearson Correlation -,344** ,171 ,155 ,142 ,064 1
N 59 59 59 59 59 59 *. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).
Table 7 Spearman’s rho
Correlations Performance_ financial Board_nr of commite es Board_ evaluations Board_ nr of meetin gs Board_ nr of membe rs Spearman' s rho SIZE Correlation 1,000 -,032 -,292* -,179 -,033 -,295* Sig. (2-tailed) . ,808 ,025 ,174 ,806 ,023 N 59 59 59 59 59 59 Performan ce_ financial Correlation -,032 1,000 ,153 ,166 ,052 ,132 Sig. (2-tailed) ,808 . ,248 ,209 ,697 ,319 N 59 59 59 59 59 59 Board_nr of commitees Correlation -,292* ,153 1,000 ,099 ,097 ,176 Sig. (2-tailed) ,025 ,248 . ,457 ,467 ,184 N 59 59 59 59 59 59 Board_ evaluations Correlation -,179 ,166 ,099 1,000 ,254 ,123 Sig. (2-tailed) ,174 ,209 ,457 . ,053 ,355 N 59 59 59 59 59 59 Board_nr of meetings Correlation -,033 ,052 ,097 ,254 1,000 -,067 Sig. (2-tailed) ,806 ,697 ,467 ,053 . ,614 N 59 59 59 59 59 59 Board_nr of members Correlation -,295* ,132 ,176 ,123 -,067 1,000 Sig. (2-tailed) ,023 ,319 ,184 ,355 ,614 . N 59 59 59 59 59 59
*. Correlation is significant at the 0.05 level (2-tailed).
In the tables above there are a few significant correlations between variables. There is no high multicollinearity because Spearman and Pearson are both under 0,7. The 0,7 norm serves as a warning according to the paper of Yu et al.(2015). There are two significant correlations. The significant correlations are between SIZE and the variable NUMBER OF BOARD
COMMITTEES. This makes sense because the bigger the company, the more committees on the board. The other significant correlation is SIZE and NUMBER OF BOARD MEMBERS. This is also logical because the bigger the company, the higher the amount of members will be. However the two significant correlations are correlation under the norm of 0,7, thereby these variables could still be included in the regression analysis.
5.4 Regression analysis
In this part of the thesis the regression analysis will be made. First I will explain the single regressions of all governance variables related to the financial performance, including the control variable. Second, there is a multiple regression of all governance variables regarding the financial performance of hospitals also the control variable is included.
5.4.1 Single regressions
At first I looked at the effect of all the governance variables on their own related to the financial performance with only the control variable size included. What is the effect of, for example, the number of board committees on the total financial performance
Table 8 single regressions financial performance
Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 2,889 ,647 4,463 ,000 SIZE ,067 ,166 ,056 ,407 ,686 Board_nr of commitees ,116 ,164 ,098 ,708 ,482 1 (Constant) 2,641 ,613 4,304 ,000 SIZE ,044 ,158 ,037 ,279 ,781 Board_nr of meetings ,081 ,065 ,165 1,247 ,217 1 (Constant) 2,184 ,817 2,672 ,010 SIZE ,120 ,167 ,100 ,718 ,476 Board_nr of members ,151 ,103 ,205 1,470 ,147 1 (Constant) 2,731 ,501 5,449 ,000 SIZE ,078 ,160 ,065 ,486 ,629 Board_evaluations ,547 ,370 ,197 1,478 ,145
a. Dependent Variable: Performance_financial
effects are all positive. This means that all governance variable have a positive effect on the financial performance of a hospital. If I relate that to the hypothesis we see that for two of the 4 hypothesis this was expected. For the hypothesis of the independent supervisory board committees and the hypothesis of the existence of board evaluations. For the other two hypothesis a negative effect was expected. However, all effects are not significant, so we cannot accept one of these hypotheses. Maybe in the multiple regression, with all governance variables and the control variable included the effect will be significant.
5.4.2 Multiple Regression
In this section the multiple regression is explained. This is the regression with all the governance variables and the control variable included. The regression table of the multiple regressions is shown in the table below.
Table 9 multiple regression
Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 3,278 ,341 9,601 ,000 SIZE ,036 ,159 ,030 ,224 ,823 2 (Constant) 1,129 1,070 1,055 ,296 SIZE ,173 ,174 ,144 ,995 ,324 Board_nr of commitees ,085 ,163 ,071 ,521 ,604 Board_evaluations ,440 ,377 ,158 1,167 ,248 Board_nr of meetings ,060 ,066 ,121 ,907 ,369 Board_nr of members ,132 ,104 ,179 1,274 ,208
a. Dependent Variable: Performance_financial
In the table above the regression of the multiple regression with all governance variables included is shown. What I see is that almost all the governance variables have a less positive effect when all variables are included. Only the positive effect on the number of members is higher is this multiple regression.
However this table suggests that all governance variables have a positive effect on the financial performance of a hospital. None of them has a significant effect on the financial performance of a hospital. In the multiple regression the significance is even lower than in the single regressions. I can say that none of the governance variables have an significant influence on the total financial performance ratio. That is not what was expected beforehand. In the following part of this thesis
I will do some additional regression analysis to search for significant effects of the governance variables regarding to financial performance. The individual ratios will be separately researched as a dependent variable. The size of the hospitals will play a role in the additional regression models.
5.5 Additional regression models
5.5.1 Different hospital sizes
In this part of the thesis the additional regression models are shown. The additional regression models are about testing the hypotheses on another level. The 3 different sizes of hospitals will be individually investigated on the influence of governance quality on the financial performance. All the financial ratios will be tested individually as independent variables. The solvency, current, margin of profit, DSCR and the profitability will all be tested on whether or not any governance variable has influence on these individual financial performance ratios. At first, the different sizes of hospitals could play a role in the outcomes of the regressions which were taken in the last part (5.4). The regression analysis regarding the different hospital sizes are shown in the table below. Table 10 Regression different hospital sizes
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta Small (Constant) -,434 1,586 -,274 ,788 Board committees ,206 ,330 ,138 ,622 ,543 Evaluations ,929 ,802 ,255 1,158 ,265 Nr of meetings ,278 ,146 ,451 * 1,913 ,075 Nr. of members ,078 ,245 ,076 ,316 ,756 Medium (Constant) 5,502 1,000 5,505 ,000 Board committees -,191 ,141 -,273 -1,352 ,199 Evaluations ,894 ,318 ,681 ** 2,807 ,015 Nr of meetings -,189 ,069 -,683 ** -2,736 ,017 Nr. of members -,113 ,117 -,209 -,958 ,355 Big (Constant) 4,241 1,573 2,696 ,016 Board committees -,004 ,282 -,003 -,013 ,989 Evaluations -,049 ,796 -,015 -,062 ,951 Nr of meetings -,139 ,100 -,333 -1,390 ,184 Nr. of members ,025 ,140 ,044 ,177 ,862
a. Dependent Variable: Performance_financial * Correlation is significant at the 10% level ** Correlation is significant at the 5% level
In the table above it is clear that there are more significant results compared to the regression made with all the sizes in one dataset. In the dataset regarding to the big hospitals there are no significant effects of the governance variables related to the total financial performance. In the dataset of medium and small hospitals there are significant results. For small hospitals there is one correlation which is significant at a 10% level. This is the effect of the number of meetings of the supervisory board on the financial performances. The Standardized Coefficients Beta of this variable is 0,451. Within small hospitals, the number of meetings have a positive significant effect. In hypothesis 3 it was expected that a greater number of meetings would have a negative influence on the financial performance. For small hospitals it is the other way round.
For the medium size hospitals there are 2 correlations which are significant at a 5% level. This is the evaluations of the supervisory board and also about the number of meetings of the board. The Standardized Coefficients Beta of the evaluations variable is 0,681. So the existence of board evaluations in medium size hospitals has a significant positive effect on the total financial performance. The Standardized Coefficients Beta of the meetings variable is -0,683. The number of meetings during one year has a significant negative effect on the total financial performances. These effects were expected in hypothesis 3 and 4.
The research to the sizes of hospitals has led to three significant effects regarding the governance variables. In big hospitals there are no significant effects and in medium size hospitals are the most significant effects of governance variables. In the big hospitals it appears to be that the supervisory board has less influence on the financial performance and in the medium size hospitals it appears to be that the supervisory has the most influence on financial performance.
All the outcomes and significant results above, could be interpreted too firm. However these findings are not very hard so further research should still be done on this subject. In the restrictions and possibilities for further research chapter of this thesis I will come back to this subject. 5.5.2 Individual financial performance
In this section of the additional regressions the individual financial performance ratios will be investigated. Whether or not the governance variables have a significant influence on the individual financial performance ratios. The individual performance ratios are: solvency ratio, current ratio, margin of profit, debt service coverage ratio and profitability ratio. The analysis focusses on the complete dataset and on the individual sizes of hospitals. There are no significant effects of the governance variables related to the current ratio, the debt service coverage ratio and the margin of profit ratio. However for the solvency ratio and the profitability ratio there are significant effects
of the governance variables. The other regression analyses without significant effects is shown in the appendix of this thesis.
Table 11 additional regression solvency/profitability
Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta (Constant) 0,109 0,054 2,042 0,046 Board_nrof members 0,016 0,009 0,241 1,878 ,065*
a. Dependent Variable: Solvency
(Constant) -0,014 0,013 -1,109 0,272
Board_nrof
members 0,004 0,002 0,272 2,132 ,037**
a. Dependent Variable: Profitability * Correlation is significant at the 10% level ** Correlation is significant at the 5% level
The regression analysis give a clear significant effect of the number of board members in relation to the ratios of solvency and profitability. These significant effects are both positive. The Standardized Coefficients Beta for solvency is 0,241 and for profitability 0,272. That means the higher the number of members in a supervisory board is, has a positive influence on the solvency and profitability ratio. This is not what was expected beforehand in hypothesis 3. The expectation was a negative influence of a higher number of board members on the financial performance. In this additional regression analysis the small, medium and big hospitals are analyzed individually related to the financial performance ratios individually. The significant results are presented below. The letter “A” under all tables are giving the different titles. The first two are the margin of profit and the profitability ratio in big hospitals. The third is the current ratio in medium sized hospitals. The last two are the profitability and DSCR ratio in small hospitals.
Table 12 additional regression individual ratios/different sizes
Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) ,068 ,033 2,099 ,052 Board commitees ,006 ,006 ,194 ,958 ,352 Evaluations ,009 ,016 ,114 ,576 ,572 Nr of meetings -,006 ,002 -,583 -2,919 ,010** Nr. of members -,005 ,003 -,331 -1,612 ,127
1 (Constant) ,062 ,029 2,131 ,049
Board commitees ,003 ,005 ,132 ,663 ,517
Evaluations ,010 ,015 ,136 ,705 ,491
Nr of meetings -,006 ,002 -,635 -3,240 ,005**
Nr. of members -,003 ,003 -,226 -1,119 ,280
a. Dependent Variable: Profitability (big hospitals)
1 (Constant) 2,212 ,614 3,606 ,003
Board commitees ,043 ,087 ,120 ,497 ,628
Evaluations ,299 ,195 ,445 1,530 ,150
Nr of meetings -,059 ,042 -,417 -1,392 ,187
Nr. of members -,139 ,072 -,503 -1,925 ,076*
a. Dependent Variable: Current ratio (medium hospitals)
1 (Constant) ,042 ,022 1,853 ,087
Board commitees -,005 ,003 -,372 -1,629 ,127
Evaluations ,013 ,007 ,517 1,881 ,083*
Nr of meetings -,001 ,002 -,261 -,921 ,374
Nr. of members -,001 ,003 -,119 -,481 ,639
a. Dependent Variable: Profitability (small hospitals)
1 (Constant) -,588 1,046 -,562 ,583
Board commitees ,070 ,218 ,065 ,320 ,754
Evaluations ,656 ,529 ,251 1,238 ,235
Nr of meetings ,280 ,096 ,629 2,911 ,011**
Nr. of members -,029 ,162 -,040 -,180 ,859
a. Dependent Variable: DSCR (small hospitals) * Correlation is significant at the 10% level ** Correlation is significant at the 5% level
In the tables above the regressions of the different sizes of hospitals are worked out related to individual financial performance ratios. In these tables only the results with significance are presented.
Within the big hospitals I see two significant effects related to two different financial ratios. For the financial ratios margin of profit and profitability the number of board meetings have a correlation which is significant at 1% level. The Standardized Coefficients Beta is -0,583 for the margin of profit and -0,635 for profitability. The effect of the number of meetings of the supervisory board on these financial performance ratios is negative, which was expected in hypothesis 2.
Within the medium sized hospitals there is one significant effect. It is the effect of the number of members in a supervisory board related to the current ratio. The correlation is significant at a 10%
level. The Standardized Coefficients Beta is -0,503. The effect of the number of members of the supervisory board on the current ratio is negative, which was expected in hypothesis 3.
Within the small sized hospitals there are two significant effects: the effect of the existence of board evaluations on the profitability and the effect of the number of meetings on the debt service coverage ratio. The correlation of the existence of board evaluations is significant at a 10% level. The Standardized Coefficients Beta is 0,517. The effect of the existence of board evaluations on the profitability is positive, which was expected in hypothesis 4. The correlation of the number of meetings on the DSCR is significant at a 5% level. The Standardized Coefficients Beta is 0,629. So the effect of the number of meetings of the supervisory board on the DSCR is positive, which was not expected in hypothesis 2.
All the outcomes and significant results above, could be interpreted too firm. However these findings are not very hard so further research should still be done on this subject. In the restrictions and possibilities for further research chapter of this thesis I will come back to this subject.
6 Conclusion and discussion
In this chapter the conclusion will be formulated and also the discussion. The restrictions of the research will also being presented in this chapter. In part 6.1 the conclusion will be described. In part 6.2 the discussion of this thesis will be discussed. In the last part of the chapter the restrictions and the possibilities of further research will be explained.
6.1 Conclusion
This thesis is about the influence of governance structure related to financial performance. This thesis points out the governance structure of the supervisory board, it is not about the board of directors. There were four hypothesis to test whether or not there is an influence of the supervisory board on the financial performance of hospitals. The governance variables are both single and multiple tested by regressions analyses. The outcomes of these regressions are discussed in the previous chapter. The conclusion is based on governance variables and financial performance variables of 59 general hospitals in the Netherlands.
When the investigation only focuses on the governance variables in relation to the total financial performance, with the control variable of size included, there is no significant outcome. Those regressions show that none of the governance variables has a significant effect on the financial performance.
However, in the additional analysis there are some significant effects. For example when the sizes are investigated separately. A higher number of meetings leads to a significant positive effect on the financial performance of small hospitals but to a negative effect on financial performance of medium sized hospitals. The existence of board evaluations could lead to a better financial performance within medium sized hospitals. Regarding the individual financial performance ratios the finding is that a higher number of board members on the supervisory board could lead to a better solvency ratio and to a better profitability ratio.
Some other findings are the following ones. A higher number of meetings has a negative influence on the margin of profit ratio within the big hospital setting. It also has a negative influence on the profitability in the big hospitals setting. Another finding is that a higher number of members on the supervisory board could have a negative influence on the current ratio in the medium sized hospital setting. The existence of board evaluations could have a positive influence on the profitability ratio for small hospitals. A higher number of supervisory board meetings could have a positive influence on the debt service coverage ratio for small hospitals.
Overall, the conclusion that could be made is the following one. The number of independent supervisory board committees on the supervisory board has no influence at the financial performance. Hypothesis 1 can be rejected. When there are higher number of meetings, in general this leads to worse financial performances. However, in small hospitals a higher number of meetings leads to better financial performances. Hypothesis 2 is for most parts rejected. Only for small hospitals you could say that the hypothesis 2 can be accepted. It is clear that more research on this part is necessary. A higher number of board members only have a positive influence on the solvency ratio and profitability ratio individually, not on the total financial performance. In case of hypothesis 3 the conclusion is that the hypothesis is rejected for the total financial performance. For the ratios solvency and profitability you could say that the hypothesis can be accepted. The existence of evaluations of the supervisory board has a positive influence on the financial performance but, only in small and medium sized hospitals. Therefore hypothesis 4 is rejected for the total population. However, for small and medium hospitals you could say that hypothesis 4 can be accepted. In that case there are some significant effects to support acceptation of this hypothesis.
6.2 Discussion
Based on the results of this thesis it can be said that the results are in line with the results of the paper of for example Dey (2008). Dey (2008) mentioned that some characteristics of governance are influencing performance, in this case financial performance. As already mentioned in the introductory chapter of this thesis, Thompsen (2014) states that not very much research has been done related to governance and non-profit organizations. This thesis should contribute to that statement of Thompsen (2014). However, as in the paper of Dey (2008) there are certainly some mixed results of the governance quality relating to, in this case financial, performance.
The statement regarding the number of board committees on a supervisory board was mentioned in the paper of Dey (2008). He suggested that when there are more committees on the supervisory board this should be better for the quality of governance and thereby better for the financial performance, which is also suggested by the paper of Christensen et al. (2010). However this research showed that there was no influence of the number of committees on a supervisory board related to financial performance of a hospital. This is not what was expected but it is not surprising because in some other papers there was no influence of the number of board committees related to financial performance. This is the case in for example the papers of Hayes et al. (2004) and Bozec (2005); these papers also failed to find any connection between supervisory board committees and performance. Maybe it is eventually not the number of committees that counts,
but for example more what kind of committees. Perhaps only the audit committee has an influence on the performance.
The aspect of the number of board meetings during one year of the supervisory board was mentioned in the papers of Dey (2008) and the paper of Brick & Chidambaran (2010). Brick & Chidambaran (2010) finds that there is no impact on the return on equity when there is higher board activity. A higher number of meetings should have a negative influence on financial performance. This is what my research suggests. However, only for big and medium sized hospitals this is in fact the case. For the small hospitals a higher number of meetings has a significant positive influence, which indicates that in smaller hospitals the supervisory board should meet more in order to get a better financial performance. The question is why only small hospitals have a positive effect. It could be that the board of directors in smaller hospitals are less qualified then the board of directors in medium and big hospitals. They should be monitored more.
The aspect of the number of members on a supervisory board was expected to have a significant negative influence on the financial performance. For example the paper of Yermack (1996) suggested that smaller board sizes are considered to be more effective in attaining higher monitoring. Smaller boards have less disagreements among board members, and are likely to be more efficient and organized in carrying out board functions, than larger boards (Yermack, 1996). However, in this research the number of members on a supervisory board has a significant positive effect on certain financial ratios. For example, the solvency ratio and the profitability ratio. Larger board sizes lead to better financial performance. The paper of Abdelsalam et al. (2008) suggested this already in their research. They stated that a larger board accumulates more information regarding the factors affecting the company’s value and it allows board members to specialize. Coles et al. (2008) assume in their paper that highly diversified companies put value on larger boards. Maybe hospitals are that divers that they should have larger boards in order to get better financial performances.
The last governance aspect that is mentioned in this thesis is about the evaluation of the supervisory board. The expectation beforehand was that this has a positive influence on the financial performance. Kiel & Nichelson (2005) believe that board evaluations can help prevent corporate failures. The paper of Manichilli et al. (2007) believes that board evaluations can contribute to effective boards and improved financial performances of firms. In the end there appears to be a significant positive influence of existence of board member evaluation in relation to financial performance. However, this significant positive influence is only in the small and medium sized hospitals. For big hospitals there are no effects in case of board evaluations.
6.3 Restrictions and further research
From the conclusion there seem to be some significant effects of governance variables on (some) financial performances.
There are a few restrictions to this thesis. These restrictions are outlined below. The first restriction is that there is only one year investigated. This is the year 2015, because this is the year with the most recent published data. The reason for choosing only one year is outlined in chapter 4 of this thesis. Another restriction is that only general hospitals are investigated and not University hospitals. This is because University hospitals are more divers than general hospitals. The third restriction of this thesis is that is chosen for investigating only the supervisory board.. Since general hospitals have a two tier board structure the board of directors is also playing a role. This board of directors is not included in this investigation.
Regarding the restrictions there are some possibilities for further research relating to this subject. The sizes of the hospitals are investigated individually and together. In further research could be chosen to only investigate one size or two sizes.
Another restriction is that there is only one year included. This year is the year 2015, all data is from 2015. Both the financial results and the governance variables. For further research more data regarding the financial performance could be investigated. In that case the investigation could work with a time-lag in the research. What is the influence of governance variables out of year t regarding to the financial results in year t+1.
In this research the general hospitals in the Netherlands were investigated but there are also University hospitals. In further research these hospitals could be used to investigate the relation between governance and financial performance.
The hospitals in the Netherlands have a two tier board structure. Which means there is a supervisory board and a board of directors. Only the supervisory board was investigated, not the board of directors. For further research the board of directors could be investigated related to financial performance of a hospital.
