• No results found

Influence of sustainability and control on financial performance Master thesis

N/A
N/A
Protected

Academic year: 2021

Share "Influence of sustainability and control on financial performance Master thesis"

Copied!
36
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Influence of sustainability and control on financial performance

Master thesis

Joshua Elsenga s2211572

Ropta 67, 9202KG Drachten, the Netherlands j.elsenga@student.rug.nl

Abstract

The stakeholder theory propagates that interests of stakeholders and organizations are aligned. Stakeholders increasingly require organizations to implement sustainability practices. Therefore organizations need to integrate sustainability in their management control systems. A win-win situation is expected whereby sustainability enhances financial performance due to improved stakeholder relations. This study contributes to research in the areas of sustainability, management control and performance effects as it is underdeveloped. For this research 36 business unit managers from organizations in the Netherlands filled out a survey, which is the main research tool. In addition 10 business unit managers were interviewed by using a short semi-structured interview. The findings are that formal sustainability control has a positive effect on financial performance, while informal sustainability control has a negative impact. When sustainability disclosure is high the effect of formal sustainability control is significantly more positive. Finally, the effect of sustainability on financial performance is negative in the short-term, but positive and bi-directional in the long-term. Key words: Sustainability, Management control system, Sustainability disclosure, Financial

performance, Sustainability performance, Stakeholder theory

University of Groningen Faculty of Economics and Business

MSc Business Administration: Organizational and Management Control

Supervisor: Dr. H.J. van Elten Co-assessor: Dr. W. Kaufmann

(2)

1 Table of content Introduction ... 2 Literature Review ... 3 Stakeholder theory ... 3 Sustainability ... 4

Sustainability and financial performance ... 4

Management control systems ... 5

Disclosure of sustainability information ... 8

Conceptual model ... 9

Methodology ... 10

Data collection and sample ... 10

Measurements ... 11

Financial performance ... 11

Sustainability performance ... 12

Use of formal SMCS and informal SMCS ... 12

Degree of sustainability disclosure ... 12

Reliability ... 12

Confirmatory factor analysis ... 13

Composite reliability ... 13 Validity ... 14 Convergent validity ... 14 Discriminant validity ... 15 Data analysis ... 15 Results ... 16 Multi-Group analysis... 20 Conclusive discussion ... 22 Managerial implications ... 24

Limitations and further research ... 24

References ... 25

Appendix A: Questionnaire ... 29

Appendix B: Interview protocol ... 32

Appendix C: Abbreviations ... 33

Appendix D: SmartPLS settings ... 34

(3)

2 Introduction

The traditional perspective on corporate performance was to maximize profits. It was also the only social responsibility of the firm, thereby the firm contributed to society (Friedman, 1962). During the 1980’s the concept of sustainability gained more attention when it became the centre of political debate. Currently, it is still an important political discussion. An example is the recent Paris Climate Conference in 2015 which attempted to achieve a legally binding agreement on climate change. Gray (2001) states that instead of being the subject of political discussion, it is of paramount importance that both social and environmental accounting become part of organisational convention, custom and law. A recent example of Royal Dutch Shell demonstrates that a similar shift takes place. Shell reached a turning point in their attitude towards sustainability. The oil company experimented with solar energy in the past but quitted in 2006. Now they are reconsidering a comeback to alternative forms of energy because the mind-set of the customers has changed. Customers require other products from Shell than they produce at the moment (Van Dijk, 2016). It illustrates that the political discussion is alive among customers (and likely other stakeholders) which pressure companies to act sustainable. Even rich companies like Shell have to respond to stakeholder demands for the sake of the continuation of the business. It is clear that sustainability is an increasingly important topic for managers as it will play a key role in the future of the business landscape (Berns et al., 2009).

Eventually, alternative paradigms to profit maximization emerged like the ‘triple bottom line’ wherein financial, social and environmental performance criteria are integrated (Gond et al., 2012). In essence, there are two opposing views that specifically focus on the interplay between the factors incorporated in the triple bottom line. The first is the trade-off approach and the second is the win-win approach (Van der Byl and Slawin-winski, 2015). The trade-off approach states that organizations are in a continuous pursuit of maximizing profits and incur costs from sustainable actions. The trade-off approach complies to the shareholder theory where the business goals are separated from society (Clarkson, 1995; Friedman, 1962; Marrewijk, 2003). Sprangler et al. (2014) argue that there exists a trade-off whereby focussing on sustainability diminishes the financial performance. To illustrate, if an organization wants to invest in a solar energy park, the costs associated with the investment come at the expense of the operating income. On the other hand, the win-win approach states that sustainability enhances financial performance. This view is in line with the stakeholder theory where the elements are aligned and business goals are linked to society (Freeman, 1984). The incurred costs of investing in sustainability are minimal and organizations benefit from sustainable actions (Ameer and Othman, 2012). To elaborate on the previous illustration, the investments in a solar park are needed to fulfil sustainable stakeholder demands. In return improved relations with stakeholders will result in enhanced financial outcomes (Orlitzky et al., 2003).

(4)

3

In line with the incorporation of sustainability in control systems, Johnson and Schaltegger (2016) argue that visions and strategies of sustainability are important, but they have to be operationalized to become effective. Therefore, sustainability management control systems (SMCS) have to be used. With such control systems in place, the subject becomes part of day-to-day business. In literature, Hosoda and Suzuki (2015) argued that sustainability is acknowledged as a new management control issue, but literature on the use of SMCS is still scarce. For example, Bebbington and Thomson (2013) indicate a gap between sustainable development and management accounting. Furthermore, Arjaliès and Mundy (2013) address recent calls in literature for empirical research into the role of management control systems (MCS) in relation to the social and environmental activities undertaken by organizations. Further, the literature review by Lueg and Radlach (2016) conclude that research on SMCS remains scarce and fragmented in relation to definitions, theoretical perspectives and performance effects. Therefore the first and second research questions are:

RQ1: What is the effect of sustainability performance on financial performance? RQ2: What is the effect of sustainability management control on financial performance?

According to the stakeholder theory organizations have to communicate their results to stakeholders to reap financial benefits (Perrini and Tencati, 2006). Stakeholders are not automatically aware of the use of an appropriate SMCS by an organization as it is part of the internal practice. By making it public an organization informs stakeholders that it is concerned with sustainability. Reports combining economic, social and environmental performance emerge and the number of organizations reporting it publicly increase (O’Dwyer and Owen, 2005). A new approach to discover the effect of sustainability disclosure is used. Organizations that disclose a lot of sustainability information are compared to those that reveal less information, if at all. This method allows to look beyond static performance effects of disclosure. The third research question is:

RQ3: What is the difference between high and low degree of sustainability disclosure in the

relationship between sustainability management control and financial performance?

This paper contributes to literature by addressing calls for more research in the area of sustainability and financial performance. The literature field is still inconclusive about the financial returns of sustainability. As sustainability will play a key role in the future of organizations it will be interesting for managers to know the financial consequences of sustainability. Moreover, it contributes to the discussion whether the stakeholder theory holds or not. Secondly, a call from literature to expand the research into sustainability and management control is addressed, which is still scarce and fragmented. Thirdly, if managers are aware of the financial effect of disclosing sustainability information about internal sustainability practices, they become aware of its importance.

Literature Review Stakeholder theory

(5)

4

But what is considered as a stakeholder? According to Clarkson (1995) stakeholders are groups or individuals that have rights or interests in an organization and its activities. He further divides stakeholders in primary and secondary stakeholders. Shareholders, employees, customers, suppliers, investors, governments, communities and other public stakeholder groups are among the primary stakeholders. He states that the survival of the organization is threatened without the participation of primary stakeholders. Media and other special interest groups are among the secondary stakeholders. They influence public opinions about the organization, but they are not essential to its survival.

In relation with sustainability the stakeholder theory states that a socially responsible organization gives attention to the legitimate interests of all stakeholders and balances them (Garriga and Melé, 2004). According to Hörisch et al. (2014) there are two approaches for using the stakeholder theory in the context of sustainability. In the first approach the natural and societal environment are considered as stakeholders. In the second approach organizations are considered as stakeholders who represent sustainability interests and interpret the natural and societal environment. For this research, the focus is on the latter approach, as this research is not focused on a stakeholder analysis. Instead the focus is on the organization itself as an entity which is responsible to act sustainable, ultimately creating value for all stakeholders.

Sustainability

Historically, corporate sustainability (CS) concerned the environment, whereas corporate social responsibility (CSR) concerned social factors. Nowadays they are considered synonyms. CS and CSR are broadly defined as company activities that include social and environmental concerns in business operations and stakeholder interactions (Arjaliès and Mundy, 2013; Van Marrewijk, 2003). The current paper adopts this broad view of sustainability. However, the actual definition of sustainability remains vague and the goal to establish a one-definition-fits-all situation is not realistic (Van Marrewijk, 2003). The definition that fits the current study is adopted from Jayanti and Rajeev Gowda (2014): ‘’sustainability encompasses voluntary corporate strategies geared towards an integration of environmental, social and economic objectives into the fabric of organisational life’’ (p. 131). This definition corresponds with the principles of the triple bottom line as there is an interplay between environmental, social and economic factors.

Jayanti and Rajeev Gowda (2014) further define the individual factors that comprise the triple bottom line, which are adopted as well. Environmental factors deal with the influence that the activities of organisations have on natural resources and pollution, among others. Social factors concern the social duties of organisations and deal with issues such as poverty, inequality, health care and other societal issues. At last, the economic factors are focused towards financial perspectives and deal with the ability of organizations to survive in competitive markets.

Sustainability and financial performance

(6)

5

First there are studies which found positive relationships between sustainability and financial performance. An example is the study by Barnett and Salomon (2012) which studied 1214 US firms in the period 1998-2006. They found an inverted U-shaped relationship between corporate social performance (CSP) and financial performance. When CSP is low, financial performance is high, and when CSP is high, financial performance is high. Wagner (2010) found a positive relationship as well. He researched 2478 US firms in the period 1992-2003. Another example is the study by Moneva and Ortas (2010). They studied 230 European companies, including companies from the Netherlands, in the period 2004-2007.

Neutral links are found by Becchetti et al. (2008). They studied 1000 US listed firms in the period 1990-2003. Several financial performance indicators were used. They concluded that total sales per employee were significantly higher while return on equity and return on investment were lower. Positive as well as negative effects cancelled each other out. Another example is the study by Rennings et al. (2003) who investigate 214 European companies in the period 1996-2001. They found that environmental performance has a positive influence on stock performance while social performance has a negative effect on stock performance. Taken together sustainability has no effect on stock performance. Yu Ting and Yen-Chun (2009) found an insignificant effect of environmental sustainability on financial performance. They studied 51 European firms from several countries in the period 2001-2007.

A negative relationship was found by Mittal et al. (2008) studying 50 Indian companies. They used ‘’having a code of ethics’’ as a proxy for sustainability initiatives. Moreover, They researched organizations from a developing country. They found A negative effect of financial performance in most observed years. Another example is the research conducted by Garcia-Castro et al. (2010). In their research of 658 US based firms they found a negative effect of sustainability performance on several financial performance indicators. An overview of the literature that is discussed in this section is provided in table 1.

The previous sections illustrate that the literature is very fragmented. Differing measures are used and different effects are found in different countries. At last the existing literature reviews and meta-analyses are discussed. The meta-analyses conducted by Lu and Taylor (2016), Margolis and Walsh (2003) and Orlitky et al. (2003) conclude that more studies found a positive relationship than a negative or neutral relationship. However, Jayanti and Rajeev Gowda (2014) argued that several literature reviews including a few meta analyses still point to a very small correlation between sustainability and financial performance. As stated in the introduction, the question if sustainability pays-off is not answered. However, conforming to the stakeholder theory sustainability performance enhances financial performance, creating value for all as the interest are aligned, therefore:

H1: The relationship between sustainability performance and financial performance is positive.

Management control systems

(7)

6

Authors Sustainability Factors Financial performance Sample Period Main findings

Positive effect

Wagner (2010) Social and environmental performance

Tobin's q (market-based measure) Panel data for a set of 2478 US firms.

1992-2003 There is a positive association between sustainability performance and economic performance. Moreover, the effect of sustainability performance is largest for high advertising intensity levels .

Moneva and Ortas (2010)

Social and environmental performance

Return on assets, Profit margin, Return on equity, Cash-flow, Operating profits

230 European firms 2004-2007 Evidence from different European countries and industries show a positive link between CSR and financial performance using a partial least squares (PLS) model.

Barnett and Salomon (2012)

Social and environmental performance

Return on assets, Net income 1214 firms and 4730 firm-year observations from the US S&P 500 and Russel 3000 Index

1998-2006 They found an inverted U-shaped relationship between corporate social performance (CSP) and financial performance. When CSP is low, the financial performance is higher than organizations with moderate CSP. When CSP is high, financial performance is the highest.

No effect

Rennings et al. (2003)

Social and environmental performance

Stock performance 214 European companies 1996-2001 A higher environmental sector performance has a significantly positive effect on stock performance, whereas a higher social sector performance has a negative effect. Moreover, when they are both tested on stock performance, the positive effect of environmental performance is offset by the negative effect of social performance resulting in no significant effect.

Becchetti et al. (2008)

Environmental performance and social performance

Return on equity, Return on investment, Return on capital employed, Total sales per employee

1000 US listed firms 1990-2003 There is no significant association between CSR and financial performance. Total sales per employee are significantly higher while return on equity and return on investment are lower.

Yu et al. (2009) Environmental performance Tobin's q, Return on assets, Return on equity, Return on investment

51 Eurpean companies 2001-2007 Insignificant effects of environmental performance on financial performance. They conclude that companies do not necessarily receive financial returns from green efforts.

Negative effect

Mittal et al. (2008) CSR measured by ''having a code of ethics''

Market value added, Economic value added

50 Indian companies from the S&P CNX Nifty

2001-2005 CSR initiatives negatively impact business performance for 3 of the 5 years observed. Garcia-Castro et al.

(2010)

Social and environmental performance

Tobin's q, Return on assets, Return on equity, Market value added

658 US-based firms 1991-2005 They found a negative effect of sustainability performance on financial performance. However this effect was found only for market value added.

(8)

7

A MCS is a tool that provides information to assist managers (Chenhall, 2003). Hosoda and Suzuki (2015) state that beside economic factors, environmental and social factors should be incorporated in the formal as well as informal aspects of the MCS. Formal controls comprise the written policies and procedures that direct behaviour in order to achieve the organization's goals. There is an agency relationship that involves a contract where the agent performs on behalf of the principal. On the other hand, informal controls comprise common values, beliefs, and traditions that direct the behaviour of group members (Chenhall, 2003). The informal control systems circumvent the formal control systems by propagating what is expected by individuals in the organization (Hosoda and Suzuki, 2015). However, formal control systems cannot regulate anything. Employees use the informal knowledge in particular situations to guide their own behaviour when no rules or procedures are provided (Falkenberg and Herremans, 1995). Both controls are important for ensuring a functioning MCS.

If an organization wants to embed sustainability principles in practice it needs to use SMCS (Songini and Pistoni, 2012). An SMCS assists managers in their pursuit of performance objectives and stakeholder goals. Besides, informal controls can also create awareness about sustainability in the organization (Durden, 2008). Although the link between MCS and financial performance is widely explored (Henri and Journeault, 2010), there are few studies that tested the link between formal and informal SMCS on financial performance. This is confirmed by the recent literature review conducted by Lueg and Radlach (2016). They found little empirical evidence on performance effects.

There are studies that discuss terms related to SMCS like ‘sustainability practices’ or ‘eco-control’. However, eco-control only concerns the environmental part of sustainability. For example, Henri and Journeault (2010) found no direct effect of eco-control on economic performance. They found an indirect mediation effect of SMCS on financial performance through environmental performance, mainly due to the positive link between eco-control and environmental performance. They studied 303 Canadian firms. Marcus and Blom (2011) found that environmental management systems have a positive effect on financial performance, but only when it concerns well-performing firms. When firms performed less well environmental management systems have a negative effect. They studied 497 firms from Germany and the UK from 1998-2006. Another research conducted by Tomsic et al. (2015) found a positive effect of the attitude towards sustainability (as proxy for sustainability practices) on economic performance. The evidence comes from 645 Slovenian small and medium-sized enterprises (SMEs) in 2013. A negative effect of the more vague term ‘sustainability practices’ on financial performance has been found by López et al. (2007), studying 55 firms from the Dow Jones Sustainability Index (DJSI) from 1998-2004.

Conclusively, specific research on formal and informal SMCS that relate to performance effects is lagging. Moreover, it is clearly fragmented. However, evidence on a positive link between MCS and financial performance is widely acknowledged in literature (Henri and Journeault, 2010). The stakeholder theory postulates that if the appropriate systems are in place, organizations are able to respond to stakeholder demands, creating value for all parties (Perrini and Tencati, 2006). Moreover, the indirect mediation effect of sustainability performance is supported (Henri et al., 2014; Henri and Journeault, 2010). Therefore, the following hypothesis are presented:

H2a: The relationship between the use of formal SMCS and financial performance is positive. H2b: The relationship between the use of informal SMCS and financial performance is positive. H3a: The relationship between the use of formal SMCS and financial performance through

sustainability performance is positive.

(9)

8 Disclosure of sustainability information

Disclosure of information in the annual report provides an important communication tool for organizations to inform the outside world, the stakeholders, about their internal practices. The benefits of these internal practices cannot be realised without associated disclosures (Toms, 2002). Sustainability disclosure consists of the disclosure of social, environmental and financial information (Michelon, 2011). The stakeholders monitor, control and perceive the organization by assessing information that is disclosed in annual reports.

According to Elkington (1994) business leaders must disclose information about sustainability otherwise competitive advantage flows away as customers turn to other organizations. Thus, the degree of sustainability disclosure is important to consider when linking the ‘’internal’’ sustainability concerns with financial performance. Disclosure of sustainability information potentially helps to improve image and relations with external actors, ultimately resulting in financial outcomes (Orlitzky et al., 2003).

To subscribe the argument of improved financial outcomes, Roberts (1992) found that economic performance is significantly related to corporate social disclosure levels. Wagner (2010) stated that sustainability influences economic performance less positively in firms with low levels of advertising intensity compared to firms with high levels. Qui et al. (2016) studied 629 firm-year observations of the largest companies from the London Stock Exchange in the period 2005-2009. They found a positive link between social disclosure and financial performance (market value). However, no link was found between environmental disclosure and financial performance. They concluded that the effect of social and environmental disclosures together was positively associated with financial performance. Finally, Ameer and Othman (2012) researched the top 100 sustainable global companies from 2006-2010, including companies from the Netherlands. They found that organizations which emphasize the reporting of corporate sustainability practices (social as well as environmental) have higher financial performance than organizations that do not.

In line with the evidence, it is expected that if the degree of sustainability disclosure is high, the effect of SMCS on financial performance is more positive compared to organizations with low degree of sustainability disclosure. The same counts for H5, as it is hypothesized previously that there is evidence of a potential mediation effect of sustainability performance. When organizations disclose more information, it is expected that the mediation effect will be more positive as well: H4a: The influence of formal SMCS on financial performance is more positive when the degree of

disclosure is high compared to low.

H4b: The influence of informal SMCS on financial performance is more positive when the degree of disclosure is high compared to low.

H5a: The influence of formal SMCS on financial performance through sustainability performance is more positive when the degree of disclosure is high compared to low.

(10)

9

Conceptual model

This study expands the research of Henri and Journeault (2010). Their conceptual framework is (partly) used as a foundation for this study. They investigated the direct effect of eco-control on financial performance of Canadian firms, and the mediation effect of environmental performance. In this study a broader definition of sustainability is used by focusing on the environmental and societal part of sustainability. The focus is on organizations from the Netherlands. The conceptual model is presented in figure 1. The independent variables are the use of formal SMCS and the use of informal SMCS. The dependent variable is financial performance. The mediation variable is sustainability performance. The variable degree of disclosure is added to the model and was not included in the study of Henri and Journeault (2010). However, disclosure is regarded as an important communication tool for organizations to inform stakeholders about its internal practices, that is, if information is disclosed. The sample is divided to compare groups with high and low degree of sustainability disclosure. It is hypothesized that the relationships are more positive if the degree of sustainability disclosure is high compared to low (high > low). Possible differences between the groups are tested in a Multi-Group Analysis (MGA).

Formal SMCS Informal SMCS Degree of sustainability disclosure Financial performance Sustainability performance H3a

+

H3b

+

H1

+

H2a

+

H2b

+

H4b high > low H5b high > low H4a high > low H5a high > low

(11)

10 Methodology

A theory-testing approach is used. This process is appropriate as theoretical explanations of the business phenomenon remain inconclusive (if sustainability pays-off). Thereafter a conceptual model and hypotheses are generated. The next steps are data collection, data analysis, results, discussion and conclusion, theoretical and managerial implications and finally limitations and suggestions for further research (Van Aken et al., 2012). Figure 2 provides a visual representation of the process steps. This section will elaborate on the data collection and data analysis steps.

Business Phenomenon Business Phenomenon Conceptual Model and Hypothesis Conceptual Model and Hypothesis Data Collection Data Collection Data Analysis Data

Analysis ResultsResults ImplicationsImplications

Limitations and Suggestions Limitations and Suggestions Figure 2 | Theory-testing process steps

Data collection and sample

Data are collected by primary data from the field, where an already available and tested questionnaire is used. The questionnaire requires respondents at the business unit level. Business unit managers allocate their efforts between activities that affect firm value immediately and those that affect firm value in the long term. In other words, they find themselves stuck in the middle. According to Abernethy et al. (2013) the collection of survey data is appropriate when the focus is on business unit managers in order to answer questions linked to managerial decision making. After the questionnaire is filled out by the business unit manager, a short semi-structured interview is conducted (Appendix B). The goal of these interview questions is to gather more in-depth information, assisting the questionnaire. The underlying rationale of the answers provide useful side-information in the elaboration of the findings. The data are collected by seven students. Each student had to interview approximately eight business unit managers from organizations that are based in the Netherlands. The network of the researcher was used to find suitable respondents. Besides, the researcher attempted to contact organizations via e-mail and telephone. Ultimately, data from ten business unit managers were collected.

After the data collection, the academic supervisor selected a random sample of 30 respondents from the complete dataset to ensure the anonymity of the organizations. The random sample was sent to the researcher. Four respondents that were collected by the researcher were already included in the random sample. As no complaints against privacy issues can be made when it comes to the data that were collected by the researcher, the other six interviews were added to the random sample. The final sample totals 36 respondents. Most data were collected in 2015 and a small part in 2016. All organizations have more than 100 employees and are classified as medium- to large companies. The managers work, on average, 6,23 years in their current position and 14,15 years for the organization. Most business unit managers have full profit and loss responsibility (55.6%). They report to a higher management level and have supervision of at least one employee. The organizations operate in different industries. The service industry is overrepresented covering 19.6% of the sample. The data are extracted from SPSS and are summarized in table 2, 3 and 4.

Mean Median Minimum Maximum

BU employees 852 690 1 10000

Organization employees 15749 1675 100 112000

Years in current position 6,23 5,25 0 18

Years in organization 14,15 10 0 40

(12)

11 Areas of responsibility Count %

Full profit and loss 20 55,6

Cost 7 19,4

Budget (not-for-profits) 9 25

Total 36 100

Table 3 | Count and distribution of the areas of responsibility

Industry Count %

Agriculture, forestry, and fishing 1 2,8

Construction 4 11,1

Finance, insurance, and real estate 3 8,3

Manufacturing 6 16,7

Retail trade 3 8,3

Services 7 19,4

Transportation, communications, electric, gas, and sanitary services

5 13,9

Wholesale trade 2 5,6

Non-classifiable establishments 4 11,1

Missing value 1 2,8

Total 36 100

Table 4 | Industry count and distribution

Measurements

The measurements in the questionnaire are adopted from previous studies (Choi and Yu, 2014; Kruis, 2008; Okongwu et al., 2013; Pondeville et al., 2013). When all the data were collected, the students had to do a data request to the supervisor as only parts of the questionnaire were needed for each individual research project. The parts that were requested for this study are presented in Appendix A. Before discussing the reliability and validity of the measurements, the variables with the corresponding measurement scales are presented.

Financial performance

In order to measure financial performance a measurement scale from the study of Choi and Yu (2014) is used. They use four questions and a 5-point scale to measure organizational performance. Choi and Yu (2014) regard these questions as a variation of the balanced scorecard method, which focus is on financial as well as non-financial measures (Kaplan and Norton, 1996). However, the questions used in Choi and Yu (2014) are focused on output measurement. The original 5-point scale is changed into a 7-point Likert scale in order to fit the measurement scale to this study. Moreover a fifth question is added concerning the overall financial performance (Table 5).

Table 5 | Balanced scorecard perspectives and measurement items from Choi and Yu (2014)

Kaplan and Norton (1996)

Choi and Yu (2014)

Financial perspective: ‘’Appearance to shareholders’’ Competitive advantage in sales and profit growth Customer perspective: ‘’Appearance to customers’’ Competitive advantage in brand improvement Internal business process perspective: ‘’Excellence in

internal processes’’

Competitive advantage in cost saving and efficiency Learning and growth perspective: ‘’Ability to change

and improve’’

(13)

12 Sustainability performance

The items to measure sustainability performance are adopted from Kruis (2008). She uses ten questions to measure subunit performance. However, for this study two questions were deleted. All the questions are adapted to incorporate the social part of sustainability. Ultimately, eight questions which are measured on a 7-point Likert scale comprise sustainability performance.

Use of formal SMCS and informal SMCS

In order to measure the use of formal and informal SMCS a measurement scale is adopted from Pondeville et al. (2013). They use eight questions and a 7-point Likert scale to measure the use of formal and informal environmental control systems. Again, the questions are slightly modified in order to include social aspects into the questions.

Degree of sustainability disclosure

Finally, this variable is measured by using the sustainability disclosure levels determined by Okongwu et al. (2013). Instead of a Likert-scale this variable is measured by the amount of sustainability subjects an organization as a whole discloses. In total there are seven types of sustainability disclosure subjects to choose from. The sustainability disclosure level will be higher if more subjects are selected by the business unit manager. In case a business unit manager is not aware of the subjects that are disclosed, the annual reports are checked by the researcher. Instead of determining the ‘’level of sustainability disclosure’’ as Okongwu et al. (2013) did, a total score of 0 to 7 is assigned dependent on the amount of subjects that an organization discloses. Eventually, the dataset is divided in two groups, one group with scores below 3 (low degree of sustainability disclosure) and one group with scores above 4 (high degree of sustainability disclosure).

Reliability

Reliability is ensured by several factors. Before the data collection explicit instructions by the supervisor were provided. The students were instructed to tell as little as possible to the respondents about the thesis subject. Moreover, the questions could not be sent in advance, as the answers were filled out by the researcher in the presence of the business unit manager. Fixed procedures decrease the influence of the researcher on the results (Van Aken et al., 2012). In addition multiple instruments are used, namely a questionnaire and a semi-structured interview. The individual questions to measure a variable can also be considered as different instruments. The use of multiple instruments is called triangulation and increases reliability (Van Aken et al., 2012). However, the sample size is very small (N=36). Moreover, every business unit manager represents one organization. As other members of the organization might tell another story reliability decreases.

Other factors that increase reliability are that a random sample of 30 respondents was selected by the supervisor, the data were collected by multiple researchers and the measurements were adopted from previous studies. Finally, the questionnaire aims at gathering perceptual data from business unit managers. Therefore, 7-point Likert scales are used in the survey. Lozano et al. (2008) investigate the amount of response categories on a Likert-scale in respect to reliability criteria. They conclude that the ideal number of response categories is between four and seven. Reliability decreases if less than four or more than seven categories are used.

(14)

13 0.381 0.464 -0.171 0.576 -0.368 FSMCS 1 FSMCS 2 FSMCS 3 FSMCS 4 FSMCS 5 FSMCS 6 FSMCS 7 FSMCS 8 ISMCS 1 ISMCS 2 ISMCS 3 ISMCS 4 ISMCS 5 ISMCS 6 ISMCS 7 ISMCS 8 FP 1 FP 2 FP 3 FP 5 0.851 0.802 0.864 0.485 0.440 0.658 0.781 0.537 0.879 0.775 0.912 0.879 0.810 0.561 0.837 0.798 0.874 0.828 0.858 0.461 0.854 0.717 0.680 0.461 0.878 0.774 0.636 0.823 Formal SMCS Informal SMCS Sustainability performance Financial performance SP 1 SP 2 SP 3 SP 4 SP 5 SP 6 SP 7 SP 8

Confirmatory factor analysis

Because the items that measure a certain variable are already used and tested by prior literature, the number of factors are set beforehand in the Confirmatory Factor Analysis (CFA). These factors correspond to the independent and dependent variables that are used in the conceptual model. Factors that do not load on a certain variable are deleted. The loadings need to have a value of above 0.4 which is acceptable (Wong, 2013). The indicator FP4 has a factor loading of 0.104 and is deleted. When the factor loadings are calculated again all the values are acceptable (Figure 3).

Figure 3 | Factor analysis without FP4

Composite reliability

The Likert scale is a reliable instrument if Cronbach’s Alpha exceeds the cut-off value (Van Aken et al., 2012). However, Cronbach’s Alpha is a traditional measure for the internal consistency between indicators and assumes that they are equally reliable. Composite reliability is more suitable for PLS-SEM (Partial least squares - Structural Equation Modelling) because it assumes individual reliability. Composite reliability can be used as a replacement for Cronbach’s Alpha (Hair et al., 2012; Wong, 2013). The composite reliability values exceed the critical value of 0.6 and therefore it is confirmed (Table 6).

Table 6 | Composite reliability

Composite reliability

FP 0.862

SP 0.899

Formal SMCS 0.877

(15)

14 Validity

Construct validity is high if the meaning of a concept corresponds to how it is measured (Van Aken et al., 2012). Construct validity consists of convergent validity and discriminant validity. It is confirmed if both are sufficiently high. Convergent validity is the extent to which a concept is completely covered. Answers from different items that intend to measure the same variable have to share variance (Bacharach, 1989). Discriminant validity indicates that the measurement has no components that do not fit the variable. If discriminant validity is confirmed the constructs can be differentiated from one another and do not share measurement items (Bacharach. 1989).

Internal validity is the extent to which conclusions about the directions of relationships are justified. This study is aware of the pitfall of inferred causality. Most researchers accept the sequential order of causality that causes precede the effects (Shugan, 2007). Cross-sectional studies, like the present study, exclude attention to the time element. Because this study is conducted at one moment in time the possibility to infer causality is limited. However, Rindfleisch et al. (2008) state that covariation and coherence are other markers of causality. Covariation is assessed by looking at the correlation between two variables. Coherence represents the degree to which independent and dependent variables conform to theory. Together these markers might enhance causal inference in cross-sectional research, but under strict conditions. This implicates that even if these conditions are met, it is still hard to infer causality. The consequence is that the hypotheses do not imply that B is an outcome or consequence of A. Rather it states that A and B are linked in one way or another. Due to time constraints the condition of collecting data in multiple periods is not met in this study. However, justification and judgement of possible directions of the hypotheses are explored with qualitative data. This increases the internal validity. Finally, the results of this study are hard to generalize as the sample size is small (N=36) and every business unit manager represents only one organization. Therefore the external validity decreases (Van Aken et al., 2012). The next sections provide the statistical analysis of convergent and discriminant validity.

Convergent validity

The cut-off value of the Average Variance Extracted (AVE) is 0.5 (Wong, 2013). the convergent validity is confirmed when this value is exceeded. It turns out that the AVE of formal SMCS is 0.484, which is not acceptable. When FSMCS4 and FSMCS5 are deleted from the initial model the AVE becomes 0.584, which confirms the convergent validity. The values for the composite reliability calculated in table 6 are slightly changed and still acceptable (Table 7). The same counts for the factor loadings (Figure 4).

AVE AVE without FSMCS 4 and FSMCS 5 Composite reliability FP 0.613 0.615 0.863 SP 0.540 0.540 0.899 Formal SMCS 0.484 0.584 0.892 Informal SMCS 0.661 0.661 0.939

(16)

15 0.407 0.440 -0.152 0.553 -0.361 FSMCS 1 FSMCS 2 FSMCS 3 FSMCS 6 FSMCS 7 FSMCS 8 ISMCS 1 ISMCS 2 ISMCS 3 ISMCS 4 ISMCS 5 ISMCS 6 ISMCS 7 ISMCS 8 FP 1 FP 2 FP 3 FP 5 0.853 0.786 0.883 0.663 0.806 0.538 0.880 0.776 0.912 0.878 0.809 0.562 0.837 0.797 0.876 0.829 0.859 0.460 0.855 0.715 0.679 0.459 0.878 0.739 0.680 0.824 Formal SMCS Informal SMCS Sustainability performance Financial performance SP 1 SP 2 SP 3 SP 4 SP 5 SP 6 SP 7 SP 8

Figure 4 | Factor analysis without FP4, FSMCS4 and FSMCS5

Discriminant validity

The square root of each AVE score is used to check the discriminant validity. These values have to be larger than the correlation values between variables (Wong, 2013). For example 0.764 is greater than the value in the same column (0.690), but is also greater than the values in the same row (0.195 and 0.710). All values satisfy this criteria and discriminant validity is confirmed.

Table 8 | Discriminant validity

Data analysis

After the reliability and validity checks, the results can be extracted from the model. SmartPLS3 is used because PLS-SEM is able to test models when the sample size is small (Wong, 2013). The data are analysed in three phases. First, SmartPLS3 is used to conduct a multiple regression analysis on H1 and H2. Second, as hypotheses H3 and H5 include mediation effects, the bootstrap method in SmartPLS3 is used as it is able to determine potential mediation effects with certainty (Ul Hadi et al., 2016). Finally, in order to test H4 and H5, a MGA is conducted in SmartPLS3 because it is an appropriate tool to test for differences between groups with high and low degree of sustainability disclosure (DoSD). The MGA determines whether the relationships in the model differ significantly from each other when the degree of sustainability disclosure changes. If that is the case the importance of DoSD can be assessed. The results of the analyses are, if available, backed up by qualitative data. The qualitative data that are used are subjectively chosen by the researcher, although the researcher attempted to include a few quotes of each individual respondent to yield more balanced conclusions as all ten respondents are also part of the quantitative dataset.

FP SP FSMCS ISMCS

FP 0.784

SP -0.020 0.735

Formal SMCS 0.195 0.710 0.764

(17)

16 Results

This section tests the conceptual model in SmartPLS3 where the hypotheses are checked on significance. In each subsection the quantitative data analysis is backed up by qualitative data to provide more in-depth analysis of the findings. Moreover, abbreviations for the variables are used (Appendix C). An overview of the settings that are used for different methods in SmartPLS3 is provided in Appendix D.

H1: The relationship between sustainability performance and financial performance is positive.

The bootstrapping tool in SmartPLS3 is used to achieve multiple regression results (Figure 5). The path coefficient of sustainability performance (SP) and financial performance (FP) is negative (B=-0.415) and significant at the 5% level (R²=0.372, p=0.032). H1 is not supported.

B = -0.415 p = 0.032 (t = 2.150) B = 0.548 p = 0.019 (t = 2.341) B = -0.368 p = 0.055 (t = 1.921) Formal SMCS Informal SMCS Sustainability performance Financial performance

Figure 5 | Direct effects of SP, FSMCS and ISMCS on FP

The negative effect is not literally noticeable from and supported by the qualitative evidence. Only one respondent is explicitly sceptical about the use of sustainability to enhance financial returns. The manager notes that there is no return from investing in sustainability:

‘’Possibly on the long-term. (…) I am sceptical. (…) Eventually it yields something, but not as much as always pretended. (…) I look at it as a neutral investment.’’ (Respondent 2)

In contrast to the finding there is far more support for a positive relationship:

‘’We have a lot of people with disabilities at work. (...) The ''people'' part really appeals. (...) Waste goes into fermentation and bread is transformed into animal nutrition, where it is revaluated. In that sense sustainability

yields, from a financial perspective, more customers.’’ (Respondent 6)

(18)

17

Most respondents claim that the positive relationship is bi-directional, that SP and FP reinforce each other. A few quotes:

‘’Sustainability means savings in energy and emissions. Investments are definitely needed. It always pays-off.’’ (Respondent 5)

‘I think financial resources are not needed to be sustainable. Often it depends, but it can reinforce each other as well. So when you put money in it, it can influence financial performance because companies place more orders

when you have improved sustainably. It goes both ways.’’ (Respondent 7)

However, further analysis show that respondents indicate that SP yields FP, but only over the long-run. Short-term investments are needed and will diminish FP. Some respondents even deter from investing in sustainability if there is no financial return expected in the future. In that sense companies ensure a win-win approach beforehand and only act sustainably if there are financial benefits in the long-term that offset the short-term investment costs. For example:

‘’You have to invest in it first, but eventually it pays itself back in 3 to 4 years. We have made a business case of it. On the one hand it is sustainability, but on the other it is financial. Of course there is a financial benefit,

otherwise we do not invest in it.’’ (Respondent 3)

‘’If you really want to benefit from sustainability you have to invest heavily, and then you will get the return on the long-term.'' (Respondent 8)

To provide further support for the short-term negative effect on FP:

‘’That is of course a bit of the chicken or the egg dilemma. But it starts with money.’’ (Respondent 4) ‘’It still begins with money.’’ (Respondent 9)

Conclusively, the qualitative evidence point to a positive relationship between SP and FP in the long-term, where even more evidence points to a bi-directional relationship. The positive long-term relationship is not subscribed by the regression result. However, this is a cross-sectional study where the time element is neglected and only term effects are observed. This might explain the short-term negative effect of SP on FP, which is illustrated by qualitative evidence.

H2a: The relationship between the use of formal SMCS and financial performance is positive. H2b: The relationship between the use of informal SMCS and financial performance is positive.

From figure 5 the path coefficient of the use of formal SMCS on FP is positive (B=0.548). The relationship is significant at the 5% level (R²=0.372, p=0.019). H2a is supported. The path coefficient of the use of informal SMCS on FP is negative (B=-0.368). The relationship is significant at the 10% level (R²=0.372, p=0.055). H2b is not supported. The use of formal SMCS has a significant positive effect on FP, while the use of informal SMCS has a marginal significant negative effect on FP.

The qualitative evidence provide further support for these findings. Eight out of ten respondents claim that the integration of sustainability in the management control system is useful, as the following quote illustrates:

(19)

18

Furthermore respondents note that the formal control system is important to realize financial goals: ‘’First, I want to have clear where we stand, where we want to go and how to go there. That is most

important.’’ (Respondent 1)

‘’You need clear formal targets.’’ (Respondent 4)

‘’If it is about financial goals, you have to work in targets, otherwise you are not going to make it. You have to be clear.’’ (Respondent 6)

These quotes provide some explanation about the negative relationship of informal SMCS on FP as well (‘’otherwise you are not going to make it’’). In contrast to this negative finding a few respondents state that the informal control system is important for achieving financial goals:

‘’Both, but I think in the industry where we are, the hospitality industry, the informal approach works better and faster. We are emotional people who provide a service.’’ (Respondent 7)

‘’The informal structure is very important. (…) That they realize what they add and what their role is in the organization… (..) We believe in informal control systems. We steer on it.’’ (Respondent 10)

Another respondent states that the informal control system is slightly more important for achieving financial goals, but there has to be a formal structure:

‘’You need a formal structure, those are your targets. (…) But in achieving it.. being informal gains you the most. That takes place in conversations, weekly meetings with team managers, daily start-ups, asking people to come

with improvements, walking through the department, that is relatively informal.’’ (Respondent 3) It is clear that the qualitative evidence relates formal SMCS with financial goals. Some managers are very explicit about the use of formal SMCS to reach financial goals. These outcomes subscribe the positive effect of formal SMCS on FP. The negative effect of informal SMCS on FP is partly supported by qualitative evidence, despite some managers noted the importance of informal SMCS for FP as well.

H3a: The relationship between the use of formal SMCS and financial performance through sustainability performance is positive.

H3b: The relationship between the use of informal SMCS and financial performance through sustainability performance is positive.

(20)

19 p = 0.013 (t = 2.482) p = 0.013 (t = 2.473) p = 0.047 (t = 1.983) P = 0.081 (t = 1.744) Formal SMCS Informal SMCS Financial performance Sustainability performance p = 0.462 (t = 0.735)

The relationships in figure 6 change due to the mediator. Formal and informal SMCS have a positive relationship with the mediator (R²=0.606, B=0.407, p=0.013; B=0.440. p=0.013). Subsequently, the mediator therefore has no effect on FP (R²=0.143, B=-0.152, p=0.462). It turns out that the positive effects of SMCS on SP neutralizes the relationship between SP and FP. As tested earlier, the direct effect of SP on FP was significantly negative. Furthermore, the direct effects of formal and informal SMCS on FP slightly weaken compared to figure 5. However, they are still significantly related, which illustrates the perishable effect of the mediator (R²=0.143, B=0.553, p=0.047; B=-0.361, p=0.0811). Table 9 and figure 6 summarize these other findings.

Path B t-values p-values Decision

FSMCS > SP 0.407 2.482 0.013 Accepted

ISMCS > SP 0.440 2.473 0.013 Accepted

SP > FP -0.152 0.462 0.735 Rejected

FSMCS > FP 0.553 1.983 0.047 Accepted

ISMCS > FP -0.361 1.744 0.081 Accepted (at

10% level) Table 9 | Mediating effects

Figure 6 | The model with mediator involved

Qualitative evidence about potential mediation effects is lagging. However there is evidence on the positive effect of formal and informal SMCS on SP. These findings indicate that organizations both use formal and informal SMCS to achieve sustainability goals. Although managers state that the subject itself, achieving sustainability goals, is more informal:

‘’I think the whole aspect is less formal. It's more that you're going to think consciously about it with each other. And with thinking about it I mean how it will be fulfilled.’’ (Respondent 1)

‘’It should be alive in an organization. Controls should be even more driven to the informal.’’ (Respondent 6) ‘’I think that's more informal because it is difficult to nail it down here. It's more of a feeling. You have to work

towards it together.’’ (Respondent 8)

It is interesting to notice that one part of the managers state the subject is more informal, while the other part state formal sustainability control is needed for achieving sustainability goals:

1

(21)

20

‘’I do not see that it is different (for sustainability goals). This is something that counts for every goal in your enterprise. Give attention to it and something happens, when you do nothing, nothing happens.’’

(Respondent 2)

‘’I think you need formal targets to get things done. Especially if it is not present by nature.’’ (Respondent 3) An interesting finding is that both formal and informal SMCS are important for achieving SP, which is subscribed by qualitative evidence. However, hypotheses H3a and H3b are not supported as no mediation effect exists.

Multi-Group analysis

The MGA is performed to test H4 and H5. Whereas before the complete dataset is tested, the dataset is now divided into two groups. The first group, with high DoSD (N=25), is compared to the second group, with low DoSD (N=10). There is one missing value. As explained earlier, the MGA is able to determine if relationships in the model differ significantly when DoSD is high compared to low.

H4a: The influence of formal SMCS on financial performance is more positive when the degree of disclosure is high compared to low.

H4b: The influence of informal SMCS on financial performance is more positive when the degree of disclosure is high compared to low.

From the bootstrapping results for both groups, provided in figure 7, clear differences in the relationships of the model appear. Concerning the relationship from formal SMCS and FP, the path coefficient is positive and significant when the DoSD is high (B=0.791, p=0.024). When DoSD is low the path coefficient turns to an insignificant positive value (B=0.534, p=0.888). The groups differ statistically significant from each other in the MGA. H4a is supported (B=0.256, p=0.048). In other words, the relationship is significantly more positive when DoSD is high. H4b is not supported (B=0.465, p=0.837). There is no significant difference between the two groups in the relationship between informal SMCS and FP.

It turns out that the relationship between formal SMCS on FP is significantly more positive when DoSD is high. Qualitative evidence support the argument that communicating about sustainability enhances FP:

‘’If you do not communicate, it will never work. From a financial point of view you have to do that because otherwise the focus diminishes. Ultimately, we can maintain our business with it.’’ (Respondent 2) ‘’Eventually it gains money. By communicating it, which is of course free publicity, you show your good side.

Ultimately guests will order more because you're doing so well (sustainably).’’ (Respondent 7) ‘’I think that it gains financially too. Let me put it differently; if you show and believe in it sincerely, not only empty phrases, but also demonstrate that you do it, it will also help with your appeal and branding. There will

probably be a gain as well because customers appreciate it.’’ (Respondent 9)

(22)

21 B = 0.182 P = 0.394 (t = 0.852) B = 0.656 P = 0.004 (t = 2.862) B = 0.791 P = 0.024 (t = 2.251) B = -0.550 P = 0.119 (t = 1.557) Formal SMCS Informal SMCS Sustainability performance Financial performance B = -0.093 P = 0.758 (t = 0.308) B = 0.709 P = 0.029 (t = 2.179) B = 0.199 P = 0.516 (t = 0.650) B = 0.534 P = 0.888 (t = 0.141) B = -1.016 P = 0.800 (t = 0.253) Formal SMCS Informal SMCS Sustainability performance Financial performance B = 0.082 P = 0.972 (t = 0.035)

High DoSD Low DoSD

‘’It is hard to communicate. We use the resources here, thus we actually show to people what we do. (…) It sounds strange, but you never see the (financial) returns (of it), but the appeal of your organization also

determines your returns. It is not measurable.’’ (Respondent 6)

Figure 7 | Bootstrapping results and relative weights (based on t-values) of high (N=25) versus low DoSD (N=10)

H5a: The influence of formal SMCS on financial performance through sustainability performance is more positive when the degree of disclosure is high compared to low.

H5b: The influence of formal SMCS on financial performance through sustainability performance is more positive when the degree of disclosure is high compared to low.

From the findings of H3a and H3b it appears that there is no mediation effect. In addition, H5a (B=0.088, p=0.350) and H5b (B=0.079, p=0.329) are not supported as well because the indirect mediation effects do not differ significantly between the two groups in the MGA. Although most hypothesized effects do not differ significantly in the MGA, except for H4a, the bootstrapping results show there are noticeable differences (Figure 7). Two interesting shifts take place (the results have to be interpreted with caution, as they are not significant in the MGA):

 The relationship between formal SMCS and SP is positive and significant when DoSD is low (B=0.709, p=0.029), and positive and insignificant if DoSD is high (B=0.182, p=0.394).

 The relationship between informal SMCS and SP is positive and significant when DoSD is high (B=0.656, p=0.004), and positive and insignificant if DoSD is low (B=0.199, p=0.516).

(23)

22

Analysis Relationship Findings Support Explanation

Bootstrapping

H1 SP > FP Negative effect Not supported SP has a significant negative effect on FP. H2a FSMCS > FP Positive effect Supported Formal SMCS has a significant positive effect on FP. H2b ISMCS > FP Negative effect Not supported Informal SMCS has a significant negative effect on

FP. Bootstrapping

test for mediation

H3a FSMCS > SP > FP No mediation effect Not supported Significant positive effect of formal SMCS on SP, but no effect of SP on FP.

H3b ISMCS > SP > FP No mediation effect Not supported Significant positive effect of informal SMCS on SP, but no effect of SP on FP. Multi-Group Analysis and bootstrapping H4a FSMCS > FP (High vs. Low)

More positive effect when DoSD is high compared to low

Supported Formal SMCS has a positive effect on FP when DoSD is high and no effect when DoSD is low. This relationship differs significantly between groups. H4b ISMCS > FP

(High vs. Low)

No difference Not supported Insignificant negative effect of informal SMCS on FP, this applies for both groups and do not differ. Multi-Group

Analysis and bootstrapping test for mediation

H5a FSMCS > SP > FP (High vs. Low)

No difference Not supported High DoSD: significant effect of informal SMCS on SP, but no effect of SP on FP. Low DoSD: significant effect of formal SMCS on SP, but no effect of SP on FP. The groups do not differ significantly from each other when it comes to the indirect effect. H5b ISMCS > SP > FP

(High vs. Low)

No difference Not supported

Table 10 | Summary of the research findings

Conclusive discussion

This section concludes the research with a discussion and elaboration of the research findings and answers the research questions. The results are compared to prior literature and possible explanations of the findings are discussed. Thereafter theoretical and managerial implications, limitations and suggestions for future research are presented.

First of all sustainability performance has a negative impact on financial performance. The negative relationship contrasts to the evidence from existing meta-analyses, which pointed to a positive relationship (Lu and Taylor, 2016; Margolis and Walsh, 2003; Orlitky et al., 2003). As illustrated previously there exists fragmentation in the field. Jayanti and Rajeev Gowda (2014) argued that literature reviews and meta analyses still point to a very small positive correlation. This study decreases the correlation even more. Because this is a cross-sectional study only short-term effects can be observed. In the short-term there is a negative effect because organizations need money to invest in sustainability. The short-term negative impact on financial performance seems to be accepted by companies, because it is offset by a positive impact in the long-term. Over the long-run the qualitative evidence pointed to a positive and bi-directional relationship. Garcia-Castro et al. (2011) subscribe the negative effect of stakeholder management on financial performance in the short-term and the positive effect over the long-run, using a longitudinal database. The findings are in favour of the stakeholder theory, but only in the long-term. RQ1 is addressed.

RQ1: What is the effect of sustainability performance on financial performance?

(24)

23

qualitative evidence illustrated a few cases where informal controls are stated to be important for achieving financial goals, but always in combination with a formal framework. An explanation for the negative effect might be that it is not possible to achieve financial goals solely on informal controls.

In the study of Henri at al., (2014) and Henri and Journeault (2010) environmental performance mediated the relationship between eco-control and financial performance. The insignificant effect of eco-control on financial performance becomes significant due to the mediator. Although this study found that sustainability controls have a positive effect on sustainability performance, the effects are not strong enough to turn the negative direct effect of sustainability performance on financial performance around. It only causes the disappearance of the negative effect. Therefore no mediation effect of sustainability performance is discovered. An explanation of this contrasting result might be that this study incorporates environmental and social aspects of sustainability, whereas Henri and Journeault (2010) neglect the social part. RQ2 is answered.

RQ2: What is the effect of sustainability management controls on financial performance?

Finally the sample group is divided in organizations with high and low degree of sustainability disclosure. Only the relationship between formal control and financial performance is significantly different and more positive when organizations disclose more information. This result is in line with Ahmeer and Othman (2012), Orlitzky et al. (2003), Roberts (1992) and Wagner (2010). They all state that disclosure has a positive effect on financial performance. The findings show that there is no difference in the relationship between informal controls and financial performance. A possible explanation might be that it is difficult to communicate or report about the use of informal controls. Another possibility is that the mind-set of stakeholders has changed as they become increasingly sceptical due to greenwashing. Greenwashing exists when organizations communicate positively about their environmental performance, although in reality the opposite is true (Delmas and Burbano, 2011). Organizations engage in greenwashing for reputational gain. Nowadays, stakeholders want to see actual results in order to be convinced by the pretended ‘’sustainability efforts’’ of organizations. They demand viable formal data as it is easier to assess than empty phrases about sustainability that are stated in missions and visions.

The bootstrapping results for the individual models of high and low disclosure showed distinguishable differences in the relationships, although not all were significant in the MGA. When organizations disclose a lot of information the influence of formal control is positively related to financial performance. An explanation for this effect is that only formal targets and procedures are disclosed in annual reports. As stakeholders are informed of the internal sustainability practices, financial performance is enhanced due to improved relations. Moreover, informal control is positively related to sustainability performance. An explanation of this effect might be that managers do not have to focus specifically on formal targets as these are already stated in the annual report. In this way informal controls can complement formal controls (Chenhall, 2003; Hosoda & Suzuki, 2015). Sustainability goals are achieved by informal controls and financial goals by formal controls.

If organizations reveal little sustainability information only formal control is positively related with sustainability performance. An explanation might be that managers need to steer on formal controls internally to achieve sustainability performance, as no formal controls are available from the annual report. No financial effects are observed, probably because no information is available for stakeholders. Therefore they cannot assess the sustainability of the organization. RQ3 is addressed.

RQ3: What is the difference between high and low degree of sustainability disclosure in the

Referenties

GERELATEERDE DOCUMENTEN

Stefan Kuhlmann is full professor of Science, Technology and Society at the University of Twente and chairing the Department Science, Technology, and Policy Studies (STePS). Earlier

As shown in [5], the hybridization of PB and DB approaches can benefit for short prediction horizons, N ∈ (0, 1) seconds, from the accuracy of the known vehicle dynamics, and for

Meer as een leser mag miskien deur hierdie beleid van die Redaksie teleurgestel voel, maar daar sal seker nog i>Jlder werke oor vervolging, internering en

Using the stochastic parallel gradient descent algorithm the modal power coefficients and the modal phases of the transverse mode-locked (TML) laser beam were reconstructed from

Key words: Factor analysis; Nature-based tourism; National park management; Sustainability; Tourist experience; Tourist satisfaction; Critical success factors

The images in the database consist of 5,748 morphed face images, where subsets consist of: (1) morphed images without post-processing to re- move digital artifacts, (2) morphed

It is shown that estimators based on sparsely connected deep neural networks with ReLU activation function and properly chosen network architecture achieve the min- imax rates

A special goal for this edition of the workshop is to use the ten-year anniversary of the CreaRE workshop to reflect on how the landscape has changed in the decade since it first