An empirical study of corporate social responsibility in the Netherlands

1

Bachelor Thesis

An empirical study of corporate social responsibility in the Netherlands

Name: Maarten Oldenburger

Student number: 10002208

Specialization: Finance & Organization

Field: Finance

Supervisor: Ieva Sakalauskaite

Research question:

Does corporate social responsibility (CSR) increase financial firm performance in the Dutch banks & insurers sector in comparison with the service sector?

2 Statement of Originality

This document is written by Student [Maarten Oldenburger/10002208] who declares to take full responsibility for the contents of this document.

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

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

3

Inhoudsopgave

CSR and firm performance ... 7

The banking sector ... 8

3. Methodology and empirical analysis ... 9

Methodology ... 10 Control variables ... 11 Hypothesis... 12 Descriptive statistics ... 13 4 Empirical results ... 16 5 Conclusion ... 19 6 References ... 21

4

1. Introduction

Since the mindset of people and companies changes to responsible entrepreneurship and products, corporate social responsibility (CSR) is a growing area in (financial) markets. It has therefore grown exponentially in the last decades (Tsoutsoura 2004). According to De Volkskrant, the amount of households that buys Fairtrade products on a regular basis

increased, since 2011, by 13%. From 50% in 2011 to a total of 63% nowadays (De Volkskrant, 30 April 2015). Pressure from customers, employees, suppliers, governments, stakeholders and other organizations have forced companies to invest in CSR. Besides this, people attach more value to the transparency of these social and environmental aspects of companies. Therefore, the demand for CSR reports increases as well. Many companies have responded positively to these concerns by issuing CSR reports, and some companies did not

(McWilliams 2000). For example, the percentage of companies in Europe that reported CSR has grown from 51% in 2008 to 71% in 2010, and even to 73% in 2013 (KPMG Survey 2013).

All these factors affect managers’ decision making. It is not just about creating shareholder value and making money as fast as possible, as managers more and more have to take these social, ethical and environmental aspects into account.

However all these CSR factors result in higher costs. This raises the question whether implementing CSR strategies have additional financial benefits for companies or not.

Therefore it is interesting to take a closer look at the effects of CSR on the performance of firms and companies.

This thesis analyzes the question whether CSR increases financial firm performance for Dutch banks & insurers in comparison with services companies. An answer is given by implementing an empirical study on the relation between CSR and financial firm

performance (return on assets). The data used come from Dutch companies on the ‘Dutch Transparency benchmark’, this list scales many Dutch companies’ and sectors‘ CSR to a scale from 0-200. Firms on this index are chosen for the years 2010 to 2014.

The second part of this thesis concerns a literature review in which the definition and measurement of CSR will be provided. It will also discuss the relation between CSR and firm performance, and differences in its relevance in banking and services industries.

5 The third section describes the methodological approach of this thesis. A description of the methodology, the hypothesis and an explanation of the model and its control

variables is given. Also a summary of the most important data is provided. In the fourth section, an overview of the empirical results is given and discussed. Finally, in the last section, a conclusion of this thesis is given and discussed. There is also a part of this section in which improvements are listed with respect to further research.

2. Literature review

This section provides an overview of CSR. Then, according to existing literature, some important outcomes of research are given. Also there will be an elucidation on how CSR is measured and what the relation between CSR and firm performance is. Finally, the banking sector is being described and there is an explanation why this thesis focuses primarily on the banks & insurers- and the services sector.

Definition of CSR

CSR is a rather vague concept. However, the first signs of what we call the modern CSR were observed in the early 1950’s (Carroll, 1999). It was a publication of Howard R. Bowen in 1953 called: Social responsibilities of the businessman. According to Bowen around a hundred of the largest companies were the vital centers of power and decision making and the actions of these firms affected the lives of citizens in many ways (Carroll, 1999). To quote Bowen’s definition of CSR: ‘It refers to the obligations of businessmen to pursue those policies, to make those decisions, or to follow those lines of action which are desirable, in terms of the objective and values of our society.’

Today there is still not one clear definition of CSR (Dahlsrud, 2006). Although most definitions have elements in common, they differ. For example, according to Business for Social Responsibility (BSR) CSR is: ‘Achieving commercial success in ways that honor ethical values and respect people, communities and the natural environment.’ According to Wartick

6 and Cochran (1985) CSR is associated with three factors: motivating principles, behavioral processes and observable outcomes of corporate and managerial actions relating to the firms relationship with its external environment. Combining these 3 factors together into a definition results in:

A business organization’s configuration of principles of social responsibility, processes of social responsiveness, and policies, programs, and observable outcomes as they relate to the

firm’s societal relationships.

McWilliams and Siegel (2001) describe it as follows: ‘Actions that appear to further some social goods, beyond the interest of the firm that which is required by law.’ What all definitions have in common is that they all take into account and value the environment, ethical and social values, community and employee relations.

CSR measurement

CSR is difficult to measure. However, a very common and general accepted method is by using a certain reputation index. On such reputation indexes companies are being rated based on one or more dimensions of social performance, like emission reduction,

sustainability and ethicality (Cochran & Wood, 1984). The two greatest disadvantages off this method are the subjectivity and the (often) small sample sizes. The fact that such rankings are subjective may well imply that they vary highly between one observer and another. This can therefore make results unreliable. Furthermore, not all companies are taken into account, but only a limited number is included in the reputation index. This causes often (too) small sample sizes (Cochran & Wood, 1984).

In this thesis, the amount of CSR strategies companies implement, is also obtained from such an index, the Dutch transparency benchmark. The fact that this index is made by the Dutch Ministry of Economic affairs makes it less subjective. For instance, environmental and social aspects of entrepreneurship, governance & remuneration and clearness criteria. An overall score is given on how transparent and responsible the companies are (on these aspects).

7

CSR and firm performance

This section describes the relationship between CSR and firm performance. Many researchers have tried to give an answer to the question whether implementing CSR is financially beneficial for firms. Different researches provide different results. Some papers do find a relation between CSR and firm performance, whilst other don’t. This

subcomponent provides the outcomes of some of the leading researches.

Aupperle and Carroll (1985) discover a negative relationship between firm

performance and the amount of CSR. This is due to the neoclassical model in which CSR is associated with costs. These costs reduce profits, so implementing CSR strategies is costly for firms and therefore, they incur a competitive disadvantage relative to competitors.

Some studies, like Waddock & Graves (1997) and Tsoutsoura (2004) found positive relationships between CSR and firm performance. According to them, the more financially strong companies can afford to invest in long-term strategies, since they can bear the associated costs. One perfect example of a long-term strategy in CSR is for example providing services to the community and/or employees, or reducing the amount of CO2 emission. Those services may be strategically linked to a stronger public image and

reputation amongst customers. This may well imply, that more skilled workers, employees, partners and other stakeholders wish to be associated with the company and are therefore more willing to join in the company (Tsoutsoura, 2004).

They also found that such companies are often more transparent, which e.g. results in a lower chance that corruption may occur. This lower risk of negative events reduces the chance of damage on reputation and related lower payoffs (Waddock and Graves, 1997)

Arlow et all. (1982) showed, that there is no significant relation between CSR and firm performance at all, nor positive nor negative. This is due to the fact that it is not easy to see the effect of CSR. They also concluded that, there may be a lag between the time that socially responsible actions are taken and the time at which these result in economically beneficial results. So in the short run, implementing CSR is neither beneficial nor harmful for a company, but on the long run it might be beneficial.

Also Alexander and Buchholz (1978) did not find any relation between CSR and market based returns. A possible explanation might be because Fama (1970) says that in efficient stock markets new information relevant to the earnings outlook of a firm is

8 immediate reflected on the stock price. So effects, if any, associated with corporate social responsibility, whether they are positive or negative, are reflected immediately in the stock price. In their research they used a very large sample which had stock returns that were no different than the market on a risk adjusted base. Under this view the effects of the degree of social responsibility on stock prices were either non-existent or had occurred prior to 1970 (Alexander & Buchholz, 1978).

These researches all differ with respect to their outcomes. This could have several reasons. Like for instance the fact that corporate social responsibility doesn’t have a clear definition. This could cause confusion for researchers. Another reason could be the fact that it is difficult to objectively measure the amount of CSR strategies companies implement. Also, most of the reputation indexes differ. Different companies are being rated on different scales. Therefore, a researcher is often relying on one index and company sample. And finally, it is quite difficult to measure the effects of CSR strategies. Since implementing CSR strategies is a long-term strategy, it is not easy to see what effect CSR has because results can only be measured after a long period of time.

The banking sector

In our modern society banks play a major role. Almost every individual has to deal with banks. In general, banks perform 3 major functions in financing economic activities, namely: screening, monitoring and enforcement. Screening is processing information and production of prospective lender- and investment opportunities. Monitoring is assessing loans once grated or assessing investments once made. Financial intermediaries monitor how firms use the funds. And finally enforcement is the way in which the financers deal with the

borrowers when contract terms are breached or in case of default. These three key actions are all equally important (Scholtens, 2009).

Besides conventional banks, which are banks that only focus on maximizing profits (Relano, 2008), there are also social (value driven) banks. Those value driven banks aim to combine maximizing value with social aspects. Socially responsible banking becomes more and more a well-established notion in the financial world. According to Scholtens (2008) more and more banks get involved in CSR. Recently (in 2013), a total of 30 major

9 international banks has signed the Equator Principles agreement. Among these major banks are: ABN AMRO, Barclays, ING, HSBC, JP Morgan Chase and Bank of America. This

agreement supports socially-responsible development and wants to set a new standard (Yeomans, 2005). According to Lemke (1987), a positive relation between CSR and the banking context exists. She reports that the Massachusetts bank has successfully promoted 138 new bank accounts with a value of 11 million dollar, by assisting endangered animal species with donations made to the World Wildlife Fund (WWF). Banks increasingly are involved with financing economic activities that aim at sustainable development and offer microcredit to the poor (Scholtens 2008). Although, there is no proper framework for people to qualify banks in this aspect.

This raises the following question. If people reward a bank’s behavior and appreciate the amount of CSR banks implement, does it also increase financial firm performance?

The comparison between the banks & insurers- and the service sector is made because banking actually is also a (special) form of providing service. They offer a form of financial service. Therefore, it is better to compare these sectors with each other than two other sectors. According to Scholtens (2008), CSR becomes more and more important and implemented in the financial world. The services sector remains behind, therefore the hypotheses in this thesis is that the relation of CSR and ROA in the banking sector is not the same as the relation in the services sector.

3. Methodology and empirical analysis

In the first section of this part, the comparing methodology of this thesis will be explained. In the further sections the hypotheses, (control) variables and descriptive data will be explained.

10

Methodology

In this thesis data from the Dutch transparency benchmark is used. This is an index

performed by the Dutch Ministry of Economic affairs. It is a research of the qualitative and quantitative developments of corporate social responsibility among the largest companies in The Netherlands. This index is published every year and ranks the Dutch companies on a scale from 0-200 on many criteria. Their motivation for performing this index on a yearly base is because: ‘The government requires companies to be transparent about their CSR policies and activities. A good report on the CSR efforts of a company enables stakeholders to engage with such company. This gives the company the opportunity to strengthen itself based on constructive criticism of its stakeholders.’ (Criteria Transparency benchmark report, 2014).

In this thesis a dataset from 2010 to 2014 is used. This is due to the fact that the results for 2015 were not available and before 2010 the ranking scale was from 0-100 instead of 0-200. Since it is not possible to compare two different scales, the choice was made to include only the years 2010 to 2014.

The Dutch transparency benchmark does not only report about companies’ CSR. A part of the score is being determined by the amount of transparency of the company (to their stakeholder). But since being transparent is also a part of CSR, this index is extremely useful for this research.

There are two types of performance measurements for companies. Accounting-based measurements and stock market-Accounting-based measurements. Unfortunately, both measurements are not perfect and have their weaknesses. One of the weaknesses of the stock market-based performance measurements may well be that it deals with asymmetric information. And if so then such measurements may not reflect fair evaluation off investors (Scholtens, 2008). McWilliams, Siegel & Wright (2006) criticize using these measurements due to the fact that they only relate to financial stakeholders, while non-financial

stakeholders are also affected by, for example, CSR activities. For this reason accounting-based measurements are used in this thesis to measure the firm performance. According to McGuire (1988) and Scholtens (2008), accounting-based measures emphasize the firm’s historical assessments of accounting profitability that capture a wide range of performance indicators such as return on assets (ROA), asset growth, operating revenue, etc. In this thesis

11 the choice has been made to use return on assets as the dependent variable because it captures short-term profitability and reflects a good, relative performance measure of the company. Therefore the ROA of different companies are easy to compare. This variable is directly taken from the ORBIS database. To find a relation between ROA and CSR, a

regression with fixed time effects is done. The control variables used in the regression will be explained in the next section.

Control variables

To find out whether a relation exists between return on assets and corporate social

responsibility, a panel regression with fixed time effects is done. Based on earlier literature certain control variables are used. A good control variable is a variable that affects both the independent and dependent variable (Tsoutsoura, 2004). Ullmann (1985), McWilliams and Siegel (2000) and Burke et al. (1986) suggest that a regression-analysis with CSR and ROA needs 3 important control variables, namely: SIZE, RISK and INDUSTRY.

According to Burke et al. (1986) SIZE is an important control variable. They

discovered that when firms grow and become larger, more attention from stakeholders is attracted. Therefore, it is important to include a control variable which reflects SIZE.

According to them, the best way to measure size is the amount of employees working at the company, or taking the logarithm of the total assets of the company. In this thesis the natural logarithm of total assets is chosen.

According to Tsoutsoura (2004) RISK is also an important control variable. As a measurement for risk, the amount of leverage is taken. This is because debt is risky due to the fact that firms that have more debt simply have to pay more (fixed) debt payments every month. This is risky because in case of default these firms can go bankrupt. So companies that have more debt should be more careful to spend their money. Since large companies can have an higher absolute amount of leverage, the relative amount of leverage is taken. This is done using the debt/assets ratio.

Finally, it is important to make a distinction between different industries using an INDUSTRY variable. Since companies in different industries can behave differently. However, this problem is circumvented by only comparing two industries with each other. So two

12 different samples of firms from a specific industry are used. Therefore an INDUSTRY variable in this regression is not necessary. At most, a dummy variable for one of the two industries could have been included. However, this is not done because 2 regressions are compared with each other.

According to Lee (2011), current ratio is also a good control variable. It is been defined as the current assets divided by the current liabilities. Nonetheless, because a lot of these values are missing (especially in the banking sector) in the databases, this variable is not included in the regression. If this control variable is despite these missing values still included, the amount of (panel) observations would decline, which causes that the regressions will become insignificant.

Finally, the CSR variable is included in the regression. This variable gives an indication of how much CSR strategies a company is using. Combining all these variables gives a panel regression of the following sort (see Regression 1). The intercept is identified with the . The i’s indicate the companies and the t’s the time, since the regression is a panel

regression. Finally the are the regression coefficients and the is the residual. To test whether a difference between the in the different sector exists, a regression with all data is done. A dummy variable CSR*BANKING is made which denotes 1 if banking sector, and 0 otherwise. Then a regression on ROA, CSR and CSR*BANKING is done.

Regression 1

it it it t it

Hypothesis

The relationship between CSR and firm performance (ROA) could take a negative or positive relation (H1) or no cohesion at all (H0). In the literature part views of different researchers are discussed. Based on earlier literature hypotheses are being made:

 H0: The relation of CSR and firm performance does not differ in both the sectors (β1banks & insurers - β1services = 0).

13

 H1: The relation of CSR and firm performance differ in the banks & insurers sector and the service sector (β1banks & insurers - β1services ≠ 0).

Descriptive statistics

Data of companies listed on the Dutch Transparency benchmark index on the banks & insurers and services sector during the years 2010-2014 were extracted from the ORBIS, AMADEUS and DATASTREAM databases. Those contain data like: return on assets, short-term liabilities, long-short-term liabilities, total liabilities, total assets, net income, etc.

In the tables 1 to 3 an overview of those data is given. In table 1, all companies (both sectors) used in this thesis are summarized. Table 2 contains the services sector companies, and table 3 contains the banks & insurers sector companies. The total amount of companies used in this thesis is 62, of which 40 in the services sector and 22 in the banks & insurers sector. Five years of data is collected, making the total amount of observations 310, of which 200 services and 110 banks & insurers sector (see Table 1, 2 & 3 respectively).

Every company was given a number since STATA can only read numbers

(CompanyID). Not every data for each company was available, making this an unbalanced panel dataset. Because the amount of current ratio observations is (especially in the banks & insurers sector) highly absent, the choice is made not to implement this variable in the regression due to the fact that the regressions will become insignificant, an overview is however given in the tables.

As denoted in the correlation table (Table 4), a correlation exists between the variables. For example, the correlation between CSR score and SIZE is positive. This

corresponds with the research of Burke et al. (1986). They argue that larger firms implement more CSR strategies due to the fact that they attract more attention from stakeholders. The correlation between return on assets and debt assets ratio is negative. This is also not suprising and suggests that firms with a higher leverage ratio (higher leverage) have a smaller return on assets. Also a (small) correlation between LEVERAGE and CSR exists. This means that companies who have a higher debt/equity ratio implement in general a little more CSR strategies.

14 Table 1

Descriptive statistics for the whole sample

Observations Mean Std. deviation Min Max CompanyID 310 1 62 YEAR 310 2010 2014 CSR 283 97.332 61.089 0 196 ROA 229 3.714 11.154 -27.158 88.457

Total assets (millions) 267 64300 193000 0.636426 1280000

lnTotal assets 267 20.939 3.273 13.363 27.877

Current Ratio 198 2.231 3.679 .039 30.904

Debt/Assets ratio 262 0.527 0.358 0 1.800

Table 2 Descriptive statistics for the services sector

Observations Mean Std. deviation Min Max CompanyID 200 1 40 YEAR 200 2010 2014 CSR 185 82 58.091 0 194 ROA 157 4.888 13.023 -27.158 88.457

Total assets (millions) 166 1710 4340 0.636426 21000

lnTotal assets 166 19.369 2.095 13.364 23.768

Current Ratio 168 2.387 3.947 .039 30.904

15 Table 3

Descriptive statistics banks & insurers sector

Observations Mean Std. deviation Min Max CompanyID 110 41 62 YEAR 110 2010 2014 CSR 98 126.276 56.157 0 196 ROA 72 1.156 4.415 -1.832 32.093

Total assets (millions) 101 1670000 2860000 9.077470 1280000 0 lnTotal assets 101 23.52 3.230 16.021 27.877 Current Ratio 30 1.363 1.666 .13 4.87 Debt/Assets ratio 101 0.433 0.456 0 0.984 Table 4 Correlation table

CSR ROA LEVERAGE SIZE

CSR 1

ROA -0.1859 1

LEVERAGE 0.0622 -0.1348 1

SIZE 0.4452 -0.2525 0.2826 1

16

4 Empirical results

In table 5 (I, II, III & IV), the outcomes of the fixed effect regression models (see regression 1) are summarized. The significance of the outcomes is denoted with the astrixes. In table 5-I, a negative relation of -0.14 between CSR and ROA is found. However, with an alpha of 12,9%, this outcome is insignificant at a 10% significance level. This causes that we cannot say there is a proper relation between ROA and CSR in this sector, but it also does not mean that there is no relation at all. The problem is that the distinction between no effect and a very small effect is to small and therefore cannot be made. This problem also arises in table 5-II. There, a negative relation of -0.015 between ROA and CSR is found. The alpha is 25.6%, making this variable also insignificant. The fact that some of the outcomes are insignificant could have several reasons. For instance, the small sample size and the (realative) large amount of missing data. It could also be that the variables used in the regression model do not explain the model propperly. The model could suffer from onmitted variable bias.

As stated in table 5-III, a weak, negative, (10 percent) significant relation of -0.005 exists between return on assets and CSR in the banks & insurers sector. This means that a relation between CSR and ROA in the banks & insurers sector exists. Only the RISK variable is insignificant here.

A possible explanation for this negative relation could be the explanation of the research of Aupperle et Carroll (1985) in which they state that implementing CSR strategies is costly. Those costs result in a reduction of the profit. The fact that Tsoutsoura (2004) found that stakeholder and other people are willing to participate more easily in a company which implements many CSR strategies makes it possible that the combination has a

negative effect, but very small.

Besides, an other explanation for this small negative relation between CSR and ROA could be the rising amount of assets due to the extra CSR strategies. Since return on assets is defined as net income divided by total assets, a larger amount of total assets reduces the return on assets. A plausible explanation for the rise in total assets could be that investors are willing to invest easier in companies with more CSR strategies making it easier for companies to issue new shares which causes a rise in the amount of cash (which is a liquid asset).

17 Finally, another possibility could be the fall of the net income. Since net income is defined as EBITDA - interest - taxes - depreciation - amortization - other expenses, this could have 2 causes: a decrease in the income or an increase in the costs. This increase in the costs for implementing those strategies was already mentioned above, but governments could also punish (fines) or reward (tax benefits) companies based on their CO2 emission. These fines causes an lower net income and thus a lower return on assets. This could either works the way around. Due to tax benefits, the amount of tax that has to be paid could reduce, which causes a rise in net income. But the rise in costs (to implement CSR strategies) outweights the tax benefits.

In table 5 IV, the outcomes of the regression of ROA on CSR*BANKING dummy are stated. There seems to be a small difference in the β’s of both sectors, but since this variable is also insignificant, no conclusions with respect to the differences in the relation of ROA and CSR in both sectors can be given.

Table 5 I Regression on both

sectors

ROA ROA ROA ROA

CSR -0.012 (0.009) -0.015 (0.009) -0.015 (0.009) -0.014 (0.009) ln(Totalassets) -1.564 (1.645) -1.009* (0.538) Debt/assets ratio -2.568 (2.011) -2.436 (1.929) Constant 5.083*** (0.971) 37.888 (34.169) 6.923*** (1.456) 28.191*** (10.978)

Year fixed effects Yes Yes Yes Yes

N 210 206 201 210

R2 0.027 0.049 0.069 0.075

*= significant at 10% significance level **=significant at 5% significance level ***=significant at 1% significance level

18 Table 5 II

Regression on services sector

ROA ROA ROA ROA

CSR -0.015 (-0.013) -0.018 (0.013) -0.014 (0.013) -0.015 (0.013) ln(Totalassets) -1.388 (3.201) 4.400 (3.630) Debt/assets ratio -22.881*** (7.751) -28.368*** (8.962) Constant 6.132*** (1.132) 39.494*** (21.687) 19.616*** (4.495) 62.924 (68.247)

Year fixed effects Yes Yes Yes Yes

N 150 146 141 141

R2 0.016 0.060 0.037 0.002

Table 5 III

Regression on banks & insurers sector

ROA ROA ROA ROA

CSR -0.006* (0.003) -0.005* (0.003) -0.006* (0.003) -0.005* (0.003) ln(Totalassets) -1.718*** (0.323) -1.697*** (0.347) Debt/assets ratio -0.629 (0.407) -0.064 (0.348) Constant 2.062*** (0.491) 42.535*** (7.629) 2.367*** (0.522) 42.067*** (8.123)

Year fixed effects Yes Yes Yes Yes

N 60 60 60 60

19 Table 5 IV

Regression with banking CSR dummy ROA ROA CSR -0.006 (0.018) -0.015 (0.011) CSR*BANKING 0.008 (0.022) Constant 4.161*** (0.807) 4.969*** (1.017)

Year fixed effects Yes Yes

N 210 210

R2 0.019 0.009

5 Conclusion

This thesis tries to give an answer to the question whether implementing CSR strategies improves financial performance for companies in the banks & insurers sector compared to companies in the service sector. A panel regression analysis with fixed time effects has been done on 2 samples to find whether such a relation between ROA and CSR exists.

This thesis concludes, that there is a significant relation between corporate social responsibility and firm performance of -0.005 (β1) in the banks & insurers sector. So, the more CSR strategies a company in this sector implements, the lower its return on assets will be. Aupperle & Carroll (1985) give a very reasonable explanation for that: implementing CSR strategies is costly for firms. Those costs could be avoided and therefore reduce profits. Also, the fact that investors are willing to invest easier in companies with more CSR strategies, makes it easier for companies to issue new shares. This causes a rise in the amount of cash, which is a liquid asset, and that causes a decrease in the return on assets.

20 The fact that governments can give tax benefits to companies that value the

environment by reducing CO2 emission or other harmfull substances could cause a increase in the return on assets, because the lower tax payments result in an increase of the net income. An increase of net income will result in a higer return on assets. But since the costs of implementing CSR strategies overweight this benefit, the return on asset does not rise, but falls.

The relation between ROA and CSR in the services sector is also small and negative, but insignificant. This implies that (at least now) no prove of a relation between CSR and firm performance in the services sector can be given. So more research needs to be done with respect to the services sector before a conclusion can be drawn. The beta which should indicate the difference between the two sectors is also insignificant. Therefore, a difference between the two sectors cannot be determined and in this thesis the H0 cannot be rejected, and an answer to the central question cannot be given.

In possible further research on CSR more work needs to be done to the services sector. With more research in this sector, a relation between firm performance and CSR could be proved. If such a relation can be proved, a better look at the different sectors could be taken. Also the data problem had to be avoided. Quite some data was not available in the databases. In particular, data of banks were not available. Some banks keep (part of their) data secret for the public, which is a disadvantage for research with respect to the banking sector. Another limitation for research on CSR is, that CSR is not easy to measure. One is limited to certain indexes and their company samples.

21

6 References

Arlow, P. & Gannon, M.J. (1982). Social responsiveness, corporate structure and economic performance.

Aupperle, K., Carroll, A., & Hatfield, J. (1985). An empirical examination of the relationship between corporate social responsibility and profitability. Academy of

management journal.

Baron (2008). Managerial contracting and corporate social responsibility. Journal of

public economics

Cochran, P. L., & Wood, R. A. (1984). Corporate social responsibility and financial performance. Academy of management Journal,27 (1), 42-56.

Dahlsrud, A. (2008). How Corporate Social Responsibility is defined: an Analysis of 37 Definitions. Corporate Social Responsibility and Environmental Management

Friedman, M. (1970). The Social Responsibility of Business is to Increase its Profits.

KPMG survey of corporate responsibility reporting. (2013). Retrieved June 14, 2015, from:http://www.kpmg.com/global/en/issuesandinsights/articlespublications/corporate

-responsibility/pages/default.aspx

Konar, S., and Cohen, M.A. (2001). Does the market performance environmental performance? The review of economics and statistic.

Lee, S. (2011). How financial slack affects firm performance: evidence from US industrial firms. Journal of Economic Research.

22 Lioui, A. & Sharma. (2012). Environmental corporate social responsibility and financial performance: Disentangling direct and indirect effects. Ecological Economics.

Margolis, J.D., Elfenbein, H.A., Walsh J.P. (2007) Does it pay to be good? A meta-analysis and redirection of research on the relationship between corporate social and financial performance.

McDonald, L.M., Rundle-Thiele, S. (2008). Corporate social responsibility and bank customer satisfaction: A research agenda. International journal of bank marketing.

McWilliams, A. Siegel & Wright (2006). Corporate social responsibility: strategic implications. Journal of management studies

Mc Williams, A., Siegel, D.A. (2000). Corporate Social Responsibility and Financial Performance: Correlation or Misspecification? Strategic management journal.

Pava, M.L., Krausz, J. (1996). The association between corporate social-responsibility and financial performance: The paradox of social cost (1996). Journal of business ethics

Relano, F. (2008). From sustainable to ethical banking. Transformation in business and

economics

Scholtens, B. (2008). A note on the interaction between corporate social responsibility and financial performance. Ecological Economics.

Scholtens, B. (2008). Corporate social responsibility in the international banking industry. Journal of business ethics.

Sen, S., Bhattacharya, C.B. (2001). Does doing good always leas to doing better?

23 Simpson, W.G., Kohers, T. (2002). The link between corporate social and financial

performance: Evidence from the banking industry. Journal of business ethics.

Stanwick, P.A., Stanwick, S.D. (1998). The Relationship Between Corporate Social Performance, and Organizational Size, Financial Performance, and Environmental Performance: An Empirical Examination. Journal of business ethics

Tsoutsoura, M. (2004). Corporate social responsibility and financial performance. Center

for responsible business UC Berkeley.

Ullman, A.A. (1985). Data in search of a theory: A critical examination of the relationship among social performance, social disclosure, and economic performance of U.S. firms.

Academy of management review.

Wartick, S.L., Cochran, P.L. (1985). The evolution of the corporate social performance model. Academy of management review.

Yeomans, M. (2005). Banks go green: the world’s largest finance firms are taking into account the environmental impact of their lending. Time

Volkskrant. (2015). Aantal huishoudens dat fairtrade koopt flink gestegen. Retrieved June 14, 2015, from: http://www.volkskrant.nl/binnenland/aantal-huishoudens-dat-fairtrade-koopt-flink-gestegen~a3990104/