MSc thesis Bram Roefs 518968

The reward for sustainable management

The effect of sustainability certification on financial performance

Author: Bram Roefs

Supervisor: dr. J.S. Small

Second reader: R. Sneep

Date: 4-6-2015

1

The reward for sustainable management

The effect of sustainability certification on financial performance

Master Thesis Strategic Management

Department of Organization and Strategy

Tilburg School of Economics and Management

Tilburg University

Author: Bram Roefs

ANR: 518968

Supervisor: dr. J.S. Small

Second reader: R. Sneep

Date: 4-6-2015

Word count: 14.610

2

Preface

This study has investigated the effect of sustainability certification on firm financial performance. As

sustainability is one of the main phenomena that today’s businesses have to deal with, it seemed like a

very interesting topic to write my thesis about.

Over the last few months I have been working many hours on this master thesis, which counts as the

final project of my studies at Tilburg University. Upon successfully completing this thesis cycle I will

obtain a master’s degree in Strategic Management.

Although it has been hard work, I am glad to say it has been a valuable experience. I would like to

take this opportunity to thank a couple of people who have supported and guided me through the

process of writing this piece of work. First of all, I would like to thank my supervisor, James Small. I

am grateful to have had him as my supervisor, as he showed genuine interest, provided constructive

criticism, valuable feedback and was always willing to help with any difficulty I encountered.

Secondly, gratitude goes out to Ruud Sneep. As the statistical analysis of my thesis proved to be quite

a challenge to me, Ruud’s help and guidance on this matter was much appreciated. Also gratitude for

his role as second reader.

Finally, a word of thanks goes out to my parents, who supported me throughout the whole process.

With his business background, my father has provided me with interesting insights and ideas how to

deal with certain things.

Bram Roefs

Tilburg, 4 June 2015

3

Management summary

With sustainability being an ever increasingly important item on the business agenda, this study

investigates whether sustainability also has an effect on the financial performance of businesses.

Specifically, the effect of sustainability certification on financial performance is studied. Certification

in this study means becoming ISO 14001 or OHSAS 18001 certified, which are certifications that

prescribe standards for sustainable management systems with regard to environment and occupational

health and safety, respectively. The central question being addressed in this study is what the effect is

of sustainability certification on financial performance for Dutch firms.

Thus far, extant literature has indicated mixed results. Though, a potentially positive relationship tends

to be favoured. Therefore, this study hypothesizes that sustainability certification has a positive effect

on financial performance for Dutch firms.

Moreover, as literature suggests that this relationship is likely to differ with firm size and

across industry types, these variables are included in this study as well. It is hypothesized that both

firm size and industry type have a moderating effect on the relationship between sustainability

certification and financial performance for Dutch firms.

Longitudinal data are collected from 65 certified Dutch firms that represent varying firm sizes and

industries. To analyse these data statistically, the random effects model is applied. Results suggest that

none of the hypotheses in this study are supported. For the main relationship, results indicate the

opposite effect; sustainability certification has a weak negative effect on financial performance for

Dutch firms. However, with industry type included in the model as a moderator, sustainability does

not seem to have a significant effect at all. Moreover, with respect to the moderating effects, no

significant results are found. Therefore, this study indicates that firm size and industry type do not

strengthen or weaken the main relationship, which is not in line with extant literature. An alternative,

descriptive approach does not result in to definite conclusions either.

So based on this study, it can be stated that the effect of sustainability certification on the financial

performance for Dutch firms is weakly negative, but not significantly when industry type is included

as moderating variable. Results for a moderating effect of firm size and industry type are inconclusive.

However, conclusions have to interpreted with caution, as this study is restricted by a number of

limitations.

4

Contents

Chapter 1 Introduction ........................................................................................................................... 6

1.1 Problem indication ....................................................................................................................... 6

1.2 Problem statement ........................................................................................................................ 8

1.3 Research questions ....................................................................................................................... 8

1.4 Research methods and data collection .......................................................................................... 9

1.5 Thesis structure ............................................................................................................................ 9

Chapter 2 Literature review and hypotheses ......................................................................................... 10

2.1 Sustainability and the business context....................................................................................... 10

2.1.1 Defining sustainability ......................................................................................................... 10

2.1.2 NGOs and governmental bodies .......................................................................................... 12

2.1.3 The business context ............................................................................................................ 12

2.1.4 Conclusion ........................................................................................................................... 14

2.2 Sustainability certification and sustainable development ........................................................... 14

2.2.1 The need for certification..................................................................................................... 14

2.2.2 Forms of certification .......................................................................................................... 15

2.2.3 Sustainable management systems ........................................................................................ 17

2.2.4 Sustainable development ..................................................................................................... 20

2.2.5 Conclusion ........................................................................................................................... 21

2.3 Sustainability certification and financial performance ................................................................ 21

2.3.1 Sustainability and financial performance ............................................................................. 21

2.3.2 Sustainability certification and financial performance ......................................................... 22

2.3.3 Conclusion ........................................................................................................................... 24

Chapter 3 Research methodology ......................................................................................................... 26

3.1 Research design .......................................................................................................................... 26

3.2 Sample strategy .......................................................................................................................... 26

3.3 Variables .................................................................................................................................... 27

3.3.1 Dependent variable .............................................................................................................. 27

3.3.2 Independent variable............................................................................................................ 28

5

3.3.3 Moderating variables ........................................................................................................... 28

3.3.4 Control variables ................................................................................................................. 28

3.4 Estimation strategy ..................................................................................................................... 29

3.5 Validity and reliability ................................................................................................................ 29

3.5.1 Validity ................................................................................................................................ 29

3.5.2 Reliability ............................................................................................................................ 30

Chapter 4 Data analysis and results ...................................................................................................... 31

4.1 Data collection............................................................................................................................ 31

4.2 The statistical model ................................................................................................................... 31

4.3 Regression output ....................................................................................................................... 32

4.4 Descriptive analysis .................................................................................................................... 36

4.5 Concluding remarks ................................................................................................................... 40

Chapter 5 Discussion ............................................................................................................................ 41

5.1 Conclusions ................................................................................................................................ 41

5.2 Discussion .................................................................................................................................. 42

5.3 Limitations and future research .................................................................................................. 43

5.4 Concluding remarks ................................................................................................................... 44

References ............................................................................................................................................ 45

Appendices ........................................................................................................................................... 50

6

Chapter 1 Introduction

In this first chapter the topic of this master thesis is introduced. Starting with a description of the

background of the topic, this chapter continues by identifying some gaps in the literature.

Subsequently, a problem statement is formulated, accompanied by the research questions. Lastly,

chapter 1 provides a short overview of the research methods and data collection, as well as the

structure of the remaining chapters of this thesis.

1.1 Problem indication

Since the World Commission on Economic Development (WCED) published its well-known report

Our Common Future in 1987, the term sustainable development has gained much popularity (Bansal,

2005). Sustainable development is concerned with ‘development that meets the needs of the present

without compromising the ability of future generations to meet their own needs’ (Brundtland, 1987). It

is a dynamic and intergenerational issue. The concept of sustainability was not a new phenomenon in

1987, but since then it has gained significantly more attention, especially in the business environment.

Until the mid-1990s, it had mainly been a societal issue, with many NGOs as its primary advocates

(Yap, 1989), but also national governments and local authorities as important actors. However, since

then, it has become an important item on the business agenda (Beckmann, Hielscher & Pies, 2014).

With the introduction of the concept of a ‘Triple Bottom Line’ in the mid 1990’s and further

popularized in 1997 by John Elkington (Norman & MacDonald, 2004), a firm’s success should not be

solely measured based on financial indicators. Instead, environmental and social/ethical performance

should be taken into account and the three components should be balanced (Norman & MacDonald,

2004; Hubbard, 2009).

Parrish (2010) argues that businesses have the means and potential to significantly contribute to

sustainable development. However, as firms need to be financially viable to survive in both the short-

and long-run, adopting sustainable business practices need to result in an economic pay-off. Some

scholars argue that sustainability can provide opportunities to establish a stronger competitive position

and long-term competitive advantage (Covin & Miles, 2000; Porter & Reinhardt, 2007). Beckmann et

al. (2014) also state that businesses can transform trade-off situations into win-win situations; this

means both sustainable development and improved financial performance instead of making a trade-

off between sustainable development and financial performance.

Other scholars are not convinced and suggest less potential for sustainable business. By

adopting a sustainable approach and going beyond legal requirements, a firm can increase its value,

generate strategic resources and insure against risks. However, a sustainable approach also implies

7

more costs, increased pressure on management-time and possibly an impairment of the relationship

between principals and agents (Wang & Bansal, 2012).

Empirical research thus far draws mixed conclusions on the effects of sustainable business operations

on the financial performance of firms (López, Garcia & Rodriguez, 2007; Hart & Ahuja, 1996; Ameer

& Othman, 2012). Therefore, it can be argued that empirical research is not conclusive yet. As

Schaltegger and Synnestvedt (2002) argue, the relationship between sustainable business practices and

financial performance might differ across countries as a result of regulations, culture and consumer

behaviour. Also, different sizes of firms and industry types might provide varying results. For

instance, many studies that have been carried out were either based on US firms from the S&P 500 or

on an international samples (i.e. Global 100, DJSI). Although studies have been found that focus on

single countries such as Spain and Germany, research in other single countries is scarce, so it is useful

to study the relationship in, for example, the Netherlands.

Another gap that this study addresses, is that previous studies have not compared different

sizes of firms. Studies were either focused on the largest firms, or samples consisted of varying firms

without comparing them. Comparing large companies with SMEs provides additional insights to the

topic. The same holds for comparing different industries. Existing studies have not explicitly

compared the effect of sustainable business practices on financial performance between different

industries.

A final contribution that this study can make is based on the measure of sustainability. Many studies

base their measure for sustainability on the reports and communication of firms. However, firms can

misuse the term sustainability and engage in practices like greenwashing and window dressing (Ramus

& Montiel, 2005). Firms use the term sustainability merely as a tool to improve its public image

instead of really embedding sustainable practices in its business. Therefore, studies might provide

misleading results when merely using company reporting. Conducting a study based on official

internationally recognized certifications on sustainability is likely to provide a more objective result.

Certification has the advantage of being externally awarded and recognized. As a result firms are

audited to ensure they follow through on the standards that these certifications entail (Bansal &

Hunter, 2003). Sustainability certifications are not minimum requirements imposed by, for instance,

governments, but they are voluntarily adopted to display proactive commitment to sustainable

development and can be seen as a strategic decision (Darnall, 2006). As a result, firms proactively

applying sustainable business practices can be distinguished from firms that merely comply with

minimum regulations. Internationally recognized sustainability certifications that are commonly

applied are ISO 14001 for the environment and OHSAS 18001 for occupational health and safety.

Investigating the effect of sustainable business practices on firm performance in this manner might be

a valuable addition to the extant literature in this field.

8

1.2 Problem statement

Based on the problem indication above, the following problem statement is formulated:

“What effect does sustainability certification have on the financial performance for Dutch firms?”

1.3 Research questions

The following research questions will be addressed in order to answer the problem statement:

Theoretical perspectives

1. How has sustainability been defined and how is it related to the business context in the

literature?

2. What forms of sustainability certification are there and how do they relate to sustainable

development?

3. According to extant theory, what relationship is there between sustainability certification and

firm financial performance and what evidence is there?

Empirical research

4. What evidence is there of a relationship between sustainability certification and financial

performance for Dutch firms?

5. Does firm size have a moderating effect on the relationship, as observed on a sample of Dutch

firms?

6. Does industry type have a moderating effect on the relationship, as observed on a sample of

Dutch firms?

Sustainability

certification

Firm size

Financial performance

Dutch firms

Industry type

9

1.4 Research methods and data collection

The approach taken in this study is to answer a set of empirical questions by making use of suitable

research methods. The primary purpose of this study is explanatory; hypotheses are developed and

tested. However, a descriptive analysis is performed as well to investigate individual trends and

particularities. The study conducted has a quantitative nature. With respect to the time horizon, this

research is longitudinal. The required quantitative data are gathered from archival databases, which are

either publically available, accessible through Tilburg University, or via personal contacts. The data

are studied by making use of statistical regression methods and descriptive analyses. Moreover,

measures are taken to safeguard the reliability and validity of this research.

1.5 Thesis structure

This thesis is divided into five chapters. The first three research questions are investigated in chapter 2,

where literature on sustainability, certification and its relationship with financial performance, the

main concepts of this research, is reviewed. Based on this literature some hypotheses are developed in

this chapter as well. Subsequently, the methodology to test these hypotheses is provided in chapter 3.

The results of testing the hypotheses and answers to research question 4, 5 and 6 are presented in

chapter 4. This chapter contains the main empirical findings of this thesis. The last research question is

answered in chapter 5. This chapter provides a discussion of the results, conclusions and implications,

directions for future research and also possible limitations of this study. Finally, references and

appendices are given at the end of the thesis.

10

Chapter 2 Literature review and hypotheses

The present chapter focusses on answering the first three research questions. This is done by providing

a review of the extant literature on the main concepts of this study. First, literature on sustainability

and its relation to the business context is discussed. Subsequently, this chapter continues with

reviewing extant literature on sustainability certification and its relation to sustainable development.

Finally, this chapter concludes with discussing the relationship between sustainability certification and

firm financial performance, as retrieved from existing literature. Based on this literature review, and

especially on the last section, hypotheses are developed that are tested in the subsequent chapters.

2.1 Sustainability and the business context

2.1.1 Defining sustainability

At the United Nations Conference on the Human Environment in 1972, the term sustainable

development was formulated for the first time (Hall, Daneke & Lenox, 2010). Initially, the meaning of

sustainable and its founding principles was quite vague and unclear. As a result, numerous and varying

definitions emerged in the field (Bansal, 2005). Definitions varied from focussing on the gap between

rich and poor countries, exploitation versus conservation of resources and intergenerational equity

(Elliott, 2006). In addition, interpretations also varied in taking a purely environmental perspective,

focussing solely on social challenges, or incorporating both aspects in one spectrum (Mawhinney,

2002).

However, it is not the purpose of this study to evaluate different interpretations of

sustainability and, with more recent history in mind, it can be stated that many scholars share a broad

consensus of what sustainability incorporates. Especially since the publication of the well-known

report Our Common Future (also known as the Brundtland-report, after the Norwegian chairman

Bruntland) in 1987 by the World Commission on Economic Development, sustainable development

gained much popularity (Bansal, 2005), and a dominant definition emerged. Following the report,

sustainable development is concerned with “development that meets the needs of the present without

compromising the ability of future generations to meet their own needs” (Brundtland, 1987). Going

beyond the current generation seems to be a common theme, as the notion of sustainability, according

to Dovers and Handmer (1992), is basically rooted in the “moral principal of intergenerational

equity”. In other words, sustainability is viewed as a dynamic phenomenon, where focus is not solely

on the current generation, but also on future generations.

For the sake of clarity, it has to be noted that the terms sustainable development and

sustainability are used interchangeably. Following Dovers and Handmer (1992), one can formally

distinguish between sustainability as a long-term end-goal and sustainable development as a

continuous process of getting closer to that end-goal. However, in this study no distinction is made

between the two terms.

11

Having identified the intergenerational nature of sustainability, we can turn to the question of how

literature has dealt with varying sustainability perspectives. Initially, sustainability was mainly viewed

from an environmental or ecological perspective. Holden, Linnerud and Banister (2014) refer to this as

‘narrow’ sustainability. However, the spectrum of sustainable development has broadened over time

and also started covering social and societal challenges. Dovers (1995) has divided the issues into

three broad categories. The first category consists of issues related to the ‘depletion and degradation’

of natural resources. Issues belonging to the second category are related to ‘pollution and waste’.

Finally, the third category includes issues of ‘society and the human condition’ (Dovers, 1995). In

accordance with these issues, Bansal (2005) argues that sustainability is about ‘environmental

integrity’, ‘social equity’ and ‘economic prosperity’. Thus, the conclusion can be drawn that

sustainability is concerned with environmental and social challenges, as well as the economic

activities they are closely tied to.

In a similar vein, but more fundamentally, Gladwin, Kennelly and Krause (1995, p.878) argue that

sustainability, although a broad concept, is based on a set of founding principles: inclusiveness,

connectivity, equity, prudence and security. They state that:

“inclusiveness implies human development over time and space. Connectivity entails an

embrace of ecological, social, and economic interdependence. Equity suggests intergenerational,

intragenerational, and interspecies fairness. Prudence connotes duties of care and prevention:

technologically, scientifically, and politically. Security demands safety from chronic threats and

protection from harmful disruption”.

These principles give a comprehensive overview of the underlying mechanisms of sustainability.

Many of the definitions used and approaches taken in recent academic literature incorporate these

mechanisms, be it sometimes implicitly. So although the literature is still lacking a precise and

objective definition of the terms sustainability and sustainable development, a holistic consensus can

be recognized in recent approaches (White, 2013). Sustainability accommodates social, environmental

and economic issues (Gimenez, Sierra & Rodon, 2012) and addressing these challenges requires an

integral approach, implying that the three aspects need to be in balance with each other (Bocken,

Short, Rana & Evans, 2014; López et al., 2007). In addition, there is a strong emphasis on the equity

within the current generation and balancing the needs of the present with future generations (Holden et

al., 2014).

12

2.1.2 NGOs and governmental bodies

Sustainability itself is not a new phenomenon, as can be concluded from the fact that the term was

already formally introduced in 1972. Even before that, individuals and organizations initiated projects

and promoted practices that can be characterized as sustainable (Bolis, Morioka & Sznelwar, 2014).

Since the 1960s, many non-governmental organizations (NGOs) have actively presented themselves as

advocates of sustainable development, both with respect to environmental and social issues (Yap,

1989). Besides NGOs, national governments and local authorities have been important actors for a

long period, by developing policy documents with various elements focussed on sustainability.

Whereas sustainability had been on the agenda of mainly NGOs and governmental bodies, the concept

gained much more and wider recognition with the publication of the Brundtland-report in 1987

(Bansal, 2005).

2.1.3 The business context

At first, sustainability propositions were met with scepticism by the business community, as they were

in conflict with the heretofore prevailing logic that economic improvement was at odds with

environmental and social sustainability (Hall, Daneke & Lenox, 2010). Following this logic, taking

into account the environment and society is likely to result in less economic growth or reduced

profitability. Hubbard (2009) argues this was in line with the dominant view in the 1980s, when the

role of a business was to exclusively create shareholder value.

Nevertheless, sustainability has become a significant item on the business agenda as this conventional

view has changed towards a wider, stakeholder-based attitude, implying that the responsibilities of the

business not only include the shareholders but also concern stakeholders such as employees,

customers, communities, and the natural environment (Steurer, 2006).

Especially with the articulation of the term ‘Triple Bottom Line’ in the mid-1990s ,which

further gained recognition in 1997 after John Elkington published his influential book entitled

‘Cannibals with Forks: The Triple Bottom Line of 21st Century Business’, sustainability has become

an important factor on the agenda (Norman & Macdonald, 2004). Henriques and Richardson (2004)

argue that the Triple Bottom Line concept has made sustainability more concrete and accessible for

the business environment. The Triple Bottom Line approach states that a firm’s performance should

not be solely evaluated based on added value in financial terms, but also by taking into account

environmental and social performance indicators. Moreover, the three aspects are of equal importance

and should be balanced (Norman & Macdonald, 2004; Hubbard, 2009; Henriques & Richardson,

2004). A variation to this, which is closely related to the Triple Bottom Line and often referred to in

the literature, is a concept that is known as the 3-P framework. In this framework the Ps stand for

People, Profits and the Planet (Miles, Munilla & Darroch, 2009).

13

This shift implies that the business environment is seen as being a primary force for sustainable

development. Subsequently, the amount of literature in this field has grown in the last two decades.

Recognition increases that conventional, unsustainable business practices, as prevalent in classical

economic theories, are, in many instances, no longer appropriate and need to be changed (López,

Garcia & Rodriguez, 2007). According to Parrish (2010) businesses have distinct means and the

potential to significantly contribute to sustainable development. However, Van Marrewijk (2003)

argues that not all firms have similar motivations for taking into account sustainability. Basically, a

business can feel an obligation to do it, can be made to do it, or sees opportunities in doing it. In many

instances, firms take sustainability into account as a result of it being imposed by increasing

stakeholder pressures, reputational interests or legal requirements and regulations. This usually

involves incremental changes and so-called end-of-pipe solutions that minimize or repair negative

impacts (Carillo-Hermosilla, del Rio & Könnölä, 2010). On the other hand, there are also businesses

that are proactive and truly embrace sustainability. In this case, businesses identify sustainability as a

significant opportunity for creating competitive advantage, increased profitability and improved

internal processes and performance. Indeed, it is argued that sustainability provides numerous

opportunities for long-term competitive advantage and a stronger competitive position (Porter &

Reinhardt, 2007; Miles & Covin, 2000). Research on the profitability of environmental investments by

Plaza-Úbeda, Burgos-Jiménez, Vazquez and Liston-Heyes (2009) suggests that such investments are

profitable in the long-term for three reasons. Besides the competitive advantage over rivals mentioned

above, environmental investments can also improve the cooperation with groups of stakeholders that

have an environmental concern. Also businesses can create or gain goodwill with, for instance,

governments.

Businesses have the ability to transform win-lose situations into win-win situations (Beckmann et al.,

2014). It has been argued that they can transcend a trade-off situation where economic gain comes at

the expense of environmental degradation or social problems; or the other way around, where

environmentally or socially sustainable practices come at the expense of lower economic growth and

profits. However, as Beckmann et al. (2014) pose, businesses are able to simultaneously generate

economic growth and environmental and social benefits. This idea is closely related to the concept of

‘creating shared value’ as articulated by Porter and Kramer (2011). Whereas businesses are often

criticized as being a substantial cause of social and environmental struggles, the advocates of the

‘shared value’ principle argue that it does not have to be like this. Although creating shared value is

still rooted in generating economic value and profits, this value needs to be generated in a way that

also produces value for society. Porter and Kramer (2011) stress that the concept of creating shared

value is not a mechanism to merely redistribute value already created by businesses. Rather, creating

shared value is concerned with enlarging the total amount of economic and societal value.

14

Taking into account the above review, it can be stated that sustainability can be seen as a source of

significant business opportunities. At the same time, and equally important, businesses contribute to a

sustainable future as well. However, from a different perspective, Wang and Bansal (2012) argue that

adopting or integrating a sustainable approach can also impose negative consequences on businesses.

Such effects can consist of more costs, increased pressure on management-time and possibly an

impairment of the relationship between principals and agents. So although many scholars, in principle,

sketch a positive view of the link between sustainability and businesses, keeping in mind possible

challenges and negative consequences remains important as well.

2.1.4 Conclusion

Reviewing the literature in this section, a number of conclusions can be drawn. Firstly, it was

identified that the term sustainability is not based on a single definition. However, extant literature

indicated a consensus with sustainability as being a concept of intra- and intergenerational equity,

where social, environmental and economic needs have to be balanced.

Thereafter, it was stated that sustainable development was initially the concern of NGOs and

governments. Only since the 1990s has it become an important consideration in the business

environment, which was the result of, for instance, the introduction of the Triple Bottom Line concept

and a gradually increasing stakeholder-based perspective.

As business has been identified as a major force for sustainable development, the body of

literature has vastly grown. Conventional, unsustainable business practices are increasingly seen as

inappropriate and need be changed. Moreover, it was argued that businesses have distinct capabilities

to contribute to sustainable development and, on the other hand, sustainability provides many new

business opportunities. Following this logic, ideas such as ‘creating shared value’ were coined,

advocating that businesses can create win-win situations, that produce value for both the business, the

environment and society.

All in all, this section has provided a comprehensive overview of sustainability and its relation

to the business context. Next section will discuss how businesses commit to sustainability, specifically

focussing on certifications, and how these relate to sustainable development.

2.2 Sustainability certification and sustainable development

2.2.1 The need for certification

As the previous section indicated, sustainability is becoming increasingly important in the business

environment. To communicate their commitment to sustainable development and create legitimacy by

showing their environmental and social performance, many businesses are publishing sustainability

reports and policy documents (O’Dwyer & Owen, 2005). Hubbard (2009) states that, in principle, this

is a positive trend. However, sustainability reporting also entails some significant concerns.

15

Sustainability reports are often developed separately from economic reports, which contradicts

the idea that sustainability is a concept where environmental, social and economic aspects are

integrated. Moreover, such reports tend to be biased and self-laudatory and do not undergo strict

auditing procedures like financial statements do (Hubbard, 2009; O’Dwyer & Owen, 2005). Indeed, as

businesses are not required by law to publish sustainability reports and policies and these documents

do not have to be audited by external third parties, it can be questioned if businesses genuinely

translate these reports and policies into actual business practices (Ramus & Montiel, 2005). Firms can

misuse the term sustainability and engage in practices like greenwashing and window dressing

(Roberts & Koeplin, 2007). This means that firms use the term sustainability merely as a tool to

improve its public image instead of really embedding sustainable practices in its business. (Ramus &

Montiel, 2005). So it can be stated that sustainability reports and policies are not completely objective

and can even be misleading with respect to consumers and investors.

With respect to objectivity, official internationally recognized sustainability certifications can

be a step in the right direction. King, Lenox and Terlaak (2005) propose that sustainability

certifications can reduce the information asymmetries that exist between businesses and external

stakeholders. As argued by Bansal and Hunter (2003), certification has the advantage of being

externally awarded and recognized. Therefore, firms are audited to ensure that they follow through on

the standards that these certifications entail. It can be argued that this ensures the credibility of

certifications and avoids the greenwashing practices described above. In addition, certificates are

commonly issued by external, independent certification bodies, which improves credibility as well

(Von Hagen, Manning & Reinecke, 2010).

Sustainability certifications are not minimum requirements imposed by, for instance,

governments, but they are voluntarily adopted practices to display proactive commitment to

sustainable development and can be seen as a strategic decision (Darnall, 2006). As a result, firms

proactively applying sustainable business practices can be properly evaluated with respect to their

sustainable performance and, moreover, can be distinguished from firms that merely comply with

minimum regulations. Von Hagen et al. (2010) conclude that sustainability certification is an

increasingly important tool for supporting sustainability.

2.2.2 Forms of certification

With certification becoming an increasingly important strategy for businesses, so is the number of

certifications increasing. Currently, there are more than 400 sustainability certifications across the

globe, all competing for adopters and public attention (Von Hagen et al., 2010). Many of these

certifications target specific industries in confined geographical areas, but there are also some

sustainability certifications that aim for global adoption and publicity (Potts, Lynch, Wilkings, Huppé,

Cunningham & Voora, 2014). Following Brunsson, Rasche and Seidl (2012), a distinction can be

made between certifications that prescribe a specific outcome and certifications that merely coordinate

16

organizational procedures without prescribing a definite outcome. The latter are more extensively

discussed in the next part on management systems. With respect to the former type, some of the most

well-known sustainability certifications are described below.

One of the first certification initiatives was the organic- or bio-standard, founded in 1972 by

the International Federation of Organic Agriculture Movements. This umbrella organization sets

minimum requirements and quality assurance systems for organic certifications (Potts et al., 2014).

Subsequently, national and local governments can build upon these minimum requirements and make

adjustments following local needs (Von Hagen et al., 2010). Businesses that meet the organic

requirements are assessed annually for recertification by external auditors (Potts et al., 2014). With

respect to the requirements, organic certification is mainly based on environmental factors and only a

few social and economic requirements are included (Von Hagen et al., 2010).

A second noted initiative is the Fairtrade Certification, finding its roots in 1988 under the

Max Havelaar label. However, the Fairtrade label as it is known today was founded in 1997 (Fairtrade

International, 2015). Fairtrade certified products are sold for premium prices, guaranteeing that small,

marginalized producers of food and agricultural products get a fixed floor price and the cooperative

they are part of gets a social premium. The aim is to empower the producers, as well as the wider

community (Potts et al., 2014; Von Hagen et al., 2010). Fairtrade International has an independent

certification branch that ensures that all products carrying the Fairtrade certification meet the required

standards and all producers are reassessed every three years (Fairtrade International, 2015). Fairtrade

has a high position on the hierarchy of sustainability certifications, and is therefore appealing to

businesses (Reinecke, Manning & Von Hagen, 2012).

Also concerned with food and agriculture is the Rainforest Alliance Certification program,

which operates in collaboration with the Network for Sustainable Agriculture. Created in 1987, the

Rainforest Alliance certification ensures that products are produced with socially and environmentally

sustainable practices (Potts et al., 2014). The Rainforest Alliance mainly promotes the conservation of

biodiversity and rainforests (Rainforest Alliance, 2015), but it also strives for the improvement of

welfare and livelihood of local communities (Von Hagen et al., 2010). As with the previous two

certifications, credibility is ensured as auditing and verification is done by external, third parties (Potts

et al., 2014).

Very similar to the Rainforest Alliance certification is the UTZ Certified initiative. Though,

whereas many certifications are introduced by social movements or NGOs, UTZ was founded by a

coffee business in 1997 (Von Hagen, 2010). UTZ certified products stand for sustainable agricultural

practices and opportunities for the producers and their community (UTZ Certified, 2015). Although,

the distinct feature of UTZ certified products is that they can be traced from producer to the shelf in

the store by a traceability system (Reinecke et al., 2012). Again, compliance with UTZ requirements is

17

closely monitored by independent external auditors and members are reassessed annually (Potts et al.,

2014).

The final, well-known, program shortly described here is the Cradle to Cradle Certification.

Introduced in 2002 and scaled up in 2010, it is a more recent initiative (C2CCertified, 2015). Cradle to

Cradle strives for regenerative product design, where at the end of the life-cycle the used materials are

turned into nutrients for the natural environment or for new products (Braungart, McDonough &

Bollinger, 2007). In other words, the aim of this initiative is to create value from waste. Products are

assessed on five quality aspects and for each aspect the product is awarded an achievement level –

basic, bronze, silver, gold, or platinum – where the lowest level represents the product’s overall

certification (C2CCertified, 2015). Certifications are administered by the Cradle to Cradle institute and

producers are audited every two year for recertification (C2CCertified, 2015). Whereas the

certifications discussed earlier are primarily focussing on developing regions, Cradle to Cradle is also

very applicable to developed, urban areas.

These five certifications are only a small example of the vast amount of certifications across the globe.

However it is possible to identify some common themes. Whether it is with a focus on the

environment, society or both, all of these initiatives promote sustainable products and production

methods. In addition, the certifications are administered by external, third parties and certified

products are subject to regular auditing by the same independent bodies. Being externally awarded

signifies credibility to consumers and reduces information asymmetry. Furthermore, it can be argued

that recognition increases as many certifications have evolved, or are evolving from small niche

markets to the mainstream public (Giovannucci & Ponte, 2005). As stated before, these certifications

involve the establishment of a specific sustainability performance standard with respect to the quality

and nature of products and services. However, such certifications do not cover the whole spectrum of

certifications; there is another type of certifications that is concerned with sustainable management

systems.

2.2.3 Sustainable management systems

Management system certifications do not include a specific outcome regarding the quality or nature of

a product (King et al., 2005), that is, they are not a performance standard that evaluates a business’

products or services (Heras-Saizarbitoria & Boiral, 2013). They are a systemized set of internal

organizational management processes and practices (Brunsson et al., 2012) and are based on a

philosophy of continuously improving these systems (Fernández-Muñiz, Montes-Peón & Vázquez-

Ordás, 2012). This can be best illustrated by means of examples.

Current leading management system certifications include ISO 14001 in the field of

environmental management systems (EMS) and OHSAS 18001 in the field of health and safety

management systems (HSMS) (King et al., 2005; Santos, Mendes & Barbosa, 2011).

18

Firms have been engaging in environmental management systems for quite some time, but ISO 14001

was the first internationally recognized EMS certification (Darnall, 2006). It was developed in 1996

and a revised edition was published in 2004 (Jørgensen, Remmen & Mellado, 2006). Currently, ISO

14001 is being revised again (ISO, 2015). In 2014 there were more than 265,000 certified businesses

(Lo, Pagell, Fan, Wiengarten & Yeung, 2014). As mentioned before, such a management system

certification is not a performance standard (Bansal & Hunter, 2012). On its website, ISO defines its

ISO 14001 standard as follows (ISO, 2015):

“ISO 14001:2004 sets out the criteria for an environmental management system and can be

certified to. It does not state requirements for environmental performance, but maps out a framework

that a company or organization can follow to set up an effective environmental management system”.

Furthermore, ISO 14001 certified management systems are based on a continuous improvement cycle;

businesses are constantly stimulated to improve their environmental management practices. This is

beneficial to businesses, as environmental concerns are becoming a fundamental element of the

business strategy (Darnall, 2006). Based on their analysis Bansal and Hunter (2003) suggest that

businesses can use ISO 14001 practices to reinforce their engagement with the natural environment.

ISO 14001 certifications can be applied on a single facility or business unit, an entire company or even

higher organizational levels (Morrow & Rondinelli, 2002).

Balzarova and Castka (2008) state that benefits from becoming ISO 14001 certified can be

plentiful. These include overall improvement of management quality, focus on less developed areas,

improved communication and interaction within a business, strengthened competitive position, but

also significant cost reductions in a number of ways (Santos et al., 2011; ISO, 2015). This is not to say

that self-developed, in-house environmental managements systems cannot provide similar value.

However, ISO 14001 has the additional benefit of being a legitimate certification, signalling to

consumers, suppliers and investors that a business is committed to environmental management (Bansal

& Hunter, 2003). Credibility is ensured, as ISO 14001 certifications are awarded externally by

independent third parties (Darnall, 2006).

The ISO 14001 standards, and EMSs in general, can be comprehensive and require additional

resources and abilities. Therefore, Tinsley (2002) argues that large companies are more likely to adopt

such standards. In reviewing research, Klewitz and Hansen (2014) conclude that only a small minority

of SMEs has an ISO 14001 certified management system in place. This does not mean that SMEs

cannot benefit from ISO 14001. It can, however, be challenging for SMEs to adopt and integrate an

environmental management system (ISO, 2015).

19

As the ISO 14001 standards for environmental management systems have gained significant

acceptance, there has also been an increasing focus on occupational health and safety management

systems; especially the OHSAS 18001 certification has come to prevail as the dominant international

standard (Fernández-Muñiz et al., 2012).

Three years after the introduction of the ISO 14001 standard, OHSAS 18001 was introduced

in 1999 (Lo et al.,2014), with the latest revised version being published in 2007. Data from 2009 show

a number of almost 60,000 OHSAS 18001 certified businesses (Lo et al., 2014). It can be stated that

the OHSAS 18001 certification is a widely applicable standard as certified businesses include

businesses of various sizes and from different industries (Chang & Liang, 2009). Fernández-Muñiz et

al. (2012) state that “the OHSAS 18001 standard aims to support and promote good practices in the

area of occupational health and safety through a systematic and structured management in order to

protect workers’ health and safety”. Similar to the ISO 14001 standard, Chang and Liang (2009) argue

that the OHSAS 18001 certification is not a performance standard; it is represented by a framework

that a business can follow to effectively recognize and control its occupational health and safety risks,

reduce the possibility for accidents and improve overall performance. The cycle of continuous

improvement is important again as well. By embedding health and safety in all layers of the business

and making use of suitable business practices, management can constantly be improved (Fernández-

Muñiz et al., 2012).

Although the OHSAS 18001 is a fairly recent standard, many firms have been actively

engaged in managing occupational health and safety for decades (Lo et al., 2014). However, as

demand increased for a recognized standard against which businesses could certify and assess their

occupational health and safety systems, the OHSAS 18001 standard has become more important

(Vinodkumar & Bhasi, 2011). This certification gives a credible signal to investors that a business has

an appropriate health and safety management system and that it is committed to this social aspect of

sustainability (Fernández-Muñiz et al., 2012). Similar to ISO 14001, businesses applying for OHSAS

18001 certification are subject to external auditing by an independent body (Fernández-Muñiz et al.,

2012). Management systems are undergo a systematic evaluation to see if business practices meet the

OHSAS 18001 standard.

Being OHSAS 18001 certified can entail a number of benefits. Lo et al. (2014) have found

that businesses adopting the OHSAS 18001 standard significantly contribute to social sustainability, as

they achieve better performance with respect to occupational health and safety than businesses that are

not certified. They better protect their human capital and simultaneously improve their public image. It

is argued that this can lead to strategic and competitive opportunities for the business as well

(Fernández-Muñiz, Montes-Peón & Vázquez-Ordás, 2009).

Although sustainable management systems are not limited to ISO 14001 and OHSAS 18001, these

two are internationally recognized and the dominant certification with respect to the environment and

20

occupational health and safety. Both share a number of similar characteristics and haven proven to

create value, both economically, environmentally and socially.

2.2.4 Sustainable development

Thus far, extant literature has indicated that business is surrounded by a vast number of certifications.

To get some sense of their contents and purposes, this study has elaborated on a small set of

internationally well-known certifications. A distinction can be made between the different types of

certifications and how they relate to sustainable development.

The first type of certifications discussed concerns standards that require a specific outcome for

a product or service; meaning that materials used have to meet certain sustainability standards, or the

processes under which they are produced have to be conform specified sustainability norms.

Giovanucci and Ponte (2005) assert that producers in developing regions capture both direct and

indirect benefits from many such sustainability certifications. Not only in direct terms of premium

pricing, but also, which are at least as important, in indirect and intangible terms, such as “the

strengthening of social capital or the improvement of community-cooperative governance structures”.

It is also argued that a number of these sustainability certifications provide necessary conditions for

the preservation of local biodiversity (Giovanucci & Ponte, 2005). Perhaps one of the most significant

certification programmes is the cradle-to-cradle standard. Whereas many certifications aim to

minimize, prevent or repair negative impacts on the environment and society, this particular standard

incorporates a positive alternative; an eco-effective approach. Braungart et al. (2007) state that cradle-

to-cradle strives for maximizing benefits for the three pillars of sustainability as well as prolonging

this value period for as long as possible. Another strength of this standard is that whereas many

certifications mainly focus on developing countries, cradle-to-cradle is applicable to developed and

developing countries alike (McDonough & Braungart, 2003). Therefore, its impact on sustainable

development can be even greater.

All in all, Giovanucci and Ponte (2005) argue that a ‘virtuous circle of empowerment’ can be

established by using the proper dynamics and endeavours to meet sustainability standards. However, it

is has to be noted that it is sometimes difficult to appropriate all the benefits, as there is a lot of

coordination, uncertainty and collective action involved.

The second type of certifications discussed in this study, the sustainable management systems, do not

directly contribute to sustainability in terms of specific products or production processes. Such

certifications map out a framework that a business can follow to integrate environmental and social

principles in its management system. Literature suggests that standards like ISO 14001 and OHSAS

18001 can embed sustainability as a core element of a business strategy (Darnall, 2006). Furthermore,

cycles of continuous improvement are fundamental for both certifications (Darnall, 2006; Fernández-

Muñiz et al., 2012), meaning that businesses are constantly reviewing, adjusting and improving their

21

sustainability awareness and performance. However, as Fernández-Muñiz et al. (2012) argue, it is not

sufficient to just implement a sustainable management system. It is of significant importance that both

top management and employees are genuinely committed to the standard in order to effectively use it.

Jørgensen (2008) also argues that as management system standards are becoming more compatible,

they can contribute further to sustainable development if firms would integrate different management

system standards into one sustainable management system.

As proposed in the literature, both types of certifications can contribute to sustainable development.

However, it requires much effort and commitment and despite an external certification, success is still

not completely guaranteed.

2.2.5 Conclusion

In this section it has been identified that sustainability certifications can be an appealing and

appropriate alternative to conventional sustainability reporting. Literature suggested that there are

many different certifications across the globe and that different types can be distinguished. The first

type of certifications discussed in this study concerned product or process certifications. Such

certifications prescribe specific sustainability standards that a product of production process has to

meet (e.g. Fairtrade and Cradle-to-cradle). The second type was characterized as sustainable

management systems (e.g. ISO 14001 and OHSAS 18001), which provide a framework that

businesses can follow to embed sustainable business practices in their management systems and

business strategy. These are less clear-cut in terms of an outcome, but provide tools for organizational

improvement.

Secondly, the relation between certifications and sustainable development was discussed.

Literature suggests that both types of certifications discussed in this study can positively contribute to

sustainable development, both directly and indirectly. Whereas this section has focussed on the

relationship between certifications and sustainable development, next section discusses the relation

between sustainability certification and financial performance of businesses.

2.3 Sustainability certification and financial performance

2.3.1 Sustainability and financial performance

Without diminishing the importance of sustainable development, sustainability practices need to result

in some economic pay-off, as firms have to be financially viable to survive in the long-run. Extant

literature includes many studies focussing on the relationship between sustainable business practices

and financial performance. However, research does not provide a univocal relationship. Indeed,

Schaltegger and Synnestvedt (2002) state that there is no natural law that proposes the relationship

between sustainability and financial performance.

Focussing on industry incumbents, Nidumolu, Prahalad and Rangaswami (2009) argue that

firms should embrace sustainability as an opportunity instead of a burden. Sustainability is likely to

22

improve performance in terms of lower costs for inputs and increased revenues from superior

products. Other studies based on established and larger firms, in general, tend to indicate a positive

relationship between sustainability and financial performance as well. With respect to environmental

sustainability, Hart and Ahuja (1996) argue that efforts to prevent pollution result in an economic pay-

off. Their results indicate that operational and financial performance of mining, manufacturing and

production firms are significantly increased after one to two years. In a similar vein it can be argued

that environmentally sustainable activities can lead to improved reputation of the firm, which

subsequently has a positive effect on financial performance (Miles & Covin, 2000). In a study based

on the top 100 sustainable global companies, Ameer and Othman (2012) find that, compared to control

firms, sustainable firms have “significant higher mean sales growth, return on assets, profit before

taxation, and cash flows from operations”.

López et al. (2007) are less convinced of a positive relationship. In a study based on European

firms, they compare firms that are included in the Dow Jones Sustainability Index with less sustainable

firms that are included in the Dow Jones Global Index. Results suggest that short-term financial

performance of sustainable firms is worse, but long-term financial performance is expected to be

superior. Wang and Bansal (2012) also argue that integrating sustainable business practices can imply

more costs, increased pressure on management-time and possibly an impairment of the relationship

between principals and agents; this negatively affects financial performance.

With respect to newly established ventures, a long-term orientation towards sustainability

practices is required to be able to benefit from sustainability. If a venture has a short-term focus,

sustainability practices are likely to lead to decreased financial performance (Wang & Bansal, 2012).

Although many studies assume that sustainability has an effect on financial performance,

Ameer and Othman (2012) have found results that question the causal direction of the relationship.

Evidence points out that there is a bi-directional relationship between sustainable business practices

and financial performance.

2.3.2 Sustainability certification and financial performance

The studies and results discussed above are based on a variety of sustainability measures. However,

empirical research focussing on the relationship between sustainability certification, specifically

management system certification, and financial performance is also present. Anecdotal evidence from

large corporations from the United States suggests that adopting the ISO 14001 certification can

significantly contribute to increased operational and financial performance (Morrow & Rondinelli,

2002). Conducting their own in-depth case studies in Germany, Morrow and Rondinelli (2002) find

that, while not the most important consequence, adopting the ISO 14001 certification is likely to have

a positive effect on cost savings and opportunities for competitive advantage. Similarly, Lo et al.

(2012) argue that, specifically in fashion and textile related industries, embracing an environmental

management system standard has a positive impact on the financial performance of firms. This is

23

mainly the result of cost savings, but also reduced lead-time and improved product quality are

important. However, other empirical studies contradict this positive view and state that the market

negatively reacts to the adoption of the ISO 14001 standard. According to Cañón-de-Francia and

Garcés-Ayerbe (2009) this negative relation holds especially for firms that are less polluting and

internationalised. Other research suggests that the relationship between environmental certification

and financial performance is insignificant. Telle (2006) concludes that ‘green’ firms, in general, tend

to perform better economically, but there is little support that this is caused by the adoption of

environmental standards.

With respect to the OHSAS 18001 certification, Fernández-Muñiz et al. (2009) conducted a

study based on Spanish construction, industrial and service firms. Results indicated that adopting this

occupational health and safety management system has a positive effect on financial performance.

Furthermore, as implemented systems are more advanced, firms are even more positive about financial

indicators. Lo et al. (2014) also support the positive relationship. Based on a sample of manufacturing

firms from the United States they have found result that suggest that OHSAS 18001 certification

“leads to significant increases in sales growth, labour productivity, and profitability”. Implementing

OHSAS 18001 has a significant and positive impact on long-run profitability and, similar to the

previous study, more advanced systems lead to increased benefits.

Although much research identifies a positive relationship between management system

certification and financial performance, the causal direction can be questioned. In a study based on

ISO 14001 certified firms in Spain, Heras-Saizarbitoria et al. (2011) find results that corroborate with

other research, by indicating that certified firms have better financial performance. However, causal

direction is ambiguous, as evidence also suggests that firms with better financial performance are

likely to have more slack resources to engage in sustainable management systems. In addition, Heras-

Saizarbitoria et al. (2011) propose a more moderate effect in that adopting a management system

certification is not a bad investment, but abnormal increases in financial performance are not likely to

occur either.

Whereas empirical research is by no means conclusive, result from the studies discussed above

tend to favour a positive relationship between sustainability certification and firm financial

performance. Therefore, the following hypothesis is proposed for the current study:

H1: Sustainability certification has a positive effect on the financial performance of Dutch

firms.

Based on extant literature and empirical research, it can be argued that the relationship differs across

industries. Vinodkumar and Bhasi (2011) argue that, in order to remain competitive, management

system certification has become specifically important for manufacturing firms. Lo et al. (2014) also

state that manufacturing firms in particular are being influenced to adopt management system

24

certifications. Likewise, Hart and Ahuja have restricted their research to mining, manufacturing and

production firms. This does not mean that the relationship exclusively holds for such industries, as

Fernández-Muñiz et al. (2009) also find positive results for service firms. Nevertheless, the effect of

sustainability certification on financial performance presumably varies across different types of

industries. Therefore, although this study suggests a positive relationship in general, the following

hypothesis is proposed:

H2: Industry type has a moderating effect on the relationship between sustainability

certification and financial performance of Dutch firms.

Similarly to the line of thought concerning industry types, it can also be argued that the relationship

differs with varying firm sizes. Many studies address the benefits and disadvantages of sustainability

certification, but most of them do not differentiate between large companies and SMEs (Heras &

Arana, 2010). As the ISO 14001 standard is comprehensive and requires additional resources and

abilities, Tinsley (2002) suggests that large firms are more likely to implement and benefit from this

certification. Fernández-Muñiz et al. (2009) support this notion and state that firm size is an important

aspect in the development of management system standards. It is argued that systems are more

extensively developed when firm size is larger. This does not mean that adopting a sustainable

management system certification is exclusively beneficial for large firms. Both large companies and

SMEs can grasp the benefits of environmental management system standards (Morrow & Rondinelli,

2002). However, SMEs can face a challenge when adopting and integrating such a certification (ISO,

2015). SMEs seem to have more problems with integrating the certified management system into daily

work routines (Heras & Arana, 2010) and they lack some of the required skills and financial resources

(Borga, Citterio, Noci & Pizzurno, 2006). Accordingly, in reviewing research, Klewitz and Hansen

(2014) find that only a small minority of SMEs has implemented an ISO 14001 certified management

system. Therefore, the current study proposes this third and final hypothesis:

H3: Firm size has a moderating effect on the relationship between sustainability certification

and financial performance of Dutch firms.

2.3.3 Conclusion

This last section has discussed some of empirical evidence that exists on the relationship between

sustainability and financial performance in general and, specifically, on the relationship between

sustainability certification and financial performance. Whereas extant literature provides arguments

for both positive and negative relationships, overall, scholars tend to favour a positive relationship.

However, it has to be noted that the relationship can be ambiguous as the causal direction is not

always clear.

25

To contribute to and extend the current literature, additional research is required. Therefore,

the first hypothesis was proposed. Furthermore, as literature indicates that extant empirical research

lacks results distinguishing between industry types and firm sizes, two additional hypotheses are

presented. The next chapter elaborates on the research methodology used to test these three

hypotheses.

26

Chapter 3 Research methodology

In the previous chapter, three hypotheses have been presented. This chapter discusses the research

methodology to test these hypotheses. First, the research design is described, followed by the sample

strategy. Subsequently, this chapter discusses the variables included in this study. Then the estimation

strategy is presented and finally the concepts of validity and reliability are discussed.

3.1 Research design

To answer the empirical research questions formulated in this study, three hypotheses are introduced

and these need to be statistically tested. If the sample size is sufficiently large, the purpose of this

study is to explain the causal relationship between sustainability certification and firm financial

performance. If the sample size is rather small, the study will have a descriptive character. To test the

hypotheses proposed in the previous chapter, measurable data are required. Therefore, the study

conducted is quantitative. As this study attempts to answer the question whether sustainability

certification has an effect on the financial performance of firms, data from multiple points in time are

necessary. This implies that, with respect to the time-horizon, this study is labelled as longitudinal.

Year 0 is the year in which a firm acquired the certification. Prior to obtaining the certification, on

average it takes between 6 and 18 months to bring the management system in practice (Lo et al.,2014).

Therefore, it is desired to obtain data for the years -2 and -1 as well. Including these years might also

indicate if a firm’s performance was already experiencing an upward or downward trend. To analyse

the effect of the certification, data for the years +1, +2 and, if applicable, +3 and +4. Data on more

years provides more insight in the long-term effect.

3.2 Sample strategy

The population of the current study consists of all ISO 14001 and OHSAS 18001 certified firms in the

Netherlands. This population can be retrieved SCCM, a Dutch institution that has a database with all

certified firms in the Netherlands. At the moment of retrieving the data from the data base, the

population consisted of 2392 firms with a valid ISO 14001 certification and 445 firms with a valid

OHSAS 18001 certification. Note that some firms have obtained both certificates, so the total

population is not the sum of all ISO 14001 and OHSAS 18001 certified firms. An overview of the

population sorted by industry can be found in appendix A. Unfortunately, the SCCM database only

provides an overview of firms that are certified; the date of initial certification is not included. This

means that, as the initial certification date is a vital component of this study, each firm has to be

analysed separately. However, due to limited time and resources, analysing all the firms from the

population is not feasible. Therefore, the firms analysed in this study are limited to the five industries

27

that include the most certified firms. Table 1a and 1b indicate the industries that are included in the

sampling procedure for ISO 14001 and OHSAS 18001 certified firms, respectively.

Table 1a: five industries with most ISO 14001 certifications

Industry Frequency %

C. 26-28 Electronics and other machinery/equipment 178 7,44%

E. Waste 196 8,19%

F. Construction 363 15,18%

H. Transport and Storage 187 7,82%

M + N. Business Services 350 14,63%

Total 1274 53,26%

Table 1b: five industries with most OHSAS 18001 certifications

Industry Frequency %

C. 19-21 Petroleum, chemical and pharmaceutical products 32 7,19%

C. 26-28 Electronics and other machinery/equipment 46 10,34%

F. Construction 42 9,44%

H. Transport and storage 48 10,79%

M + N. Business Services 56 12,58%

Total 224 50,34%

Industry codes are based on the Standaard Bedrijfs Indeling (Standard firm classification) from the

Dutch Centraal Bureau voor de Statistiek (Statistics Netherlands). For this study a non-probability

sampling technique is applied, which means that the sample will not statistically represent the

complete population. However, random sampling is not suitable, as not all firms provide the data that

are required for the analysis. Therefore, this study applies purposive sampling. First, for each of the

firms the initial certification date needs to be identified, which is done by searching on the companies’

websites. Thereafter, the required data to measure firm size and firm performance for the remaining

firms are gathered from two databases: Orbis and Creditsafe.

3.3 Variables

3.3.1 Dependent variable

The dependent variable in the conceptual model of this study is firm financial performance. Various

measures can be used to operationalize this variable. In this study, the measure return on assets (ROA)

is used to operationalize firm financial performance. ROA is a measure of operating performance and

can be defined as net income divided by total assets. As mentioned before, ROA is not the only

28

measure to operationalize firm financial performance. However, it is frequently used in similar studies,

which is likely to safeguard the reliability of the study. Data are obtained from Orbis and Creditsafe.

3.3.2 Independent variable

The independent variable in the conceptual model of this study is sustainability certification,

specifically the ISO 14001 standard for environmental management systems and the OHSAS 18001

standard for occupational health and safety management systems. This study explicitly focusses on

certification, as it is an objective measure of sustainability; certifications are internationally

recognized, externally awarded and regularly audited. As mentioned earlier, this precludes that

conclusions are drawn based on subjective and possibly misleading information that firm reporting

and communication often entails. Certification is coded as a binary variable, with value 0 at T-2 and

T-1, and value 1 at T0 and onwards.

3.3.3 Moderating variables

In the conceptual model of this study two moderating variables are included. The first moderating

variable is industry type. To operationalize industry type, dummy variables for the five most important

industries of both the certifications are generated. Codes from the Dutch SBI (Standard firm

classification) have been used to do so. As there is much overlay between the top five industries for

both certifications, the number of dummy variables is six. Dummy 1 is the base category.

The second moderating variable in this study is firm size. This construct is operationalized by

making use of three measures: number of employees, annual balance sheet total and annual turnover.

Using three measures is motivated by the fact that it gives a more complete notion, as firm size has

multiple aspects. To test for moderation, all of these variables are interacted with the independent

variable.

3.3.4 Control variables

Two variables that are frequently used as control variables, industry effects and firm size, are part of

this study explicitly as moderating variables. However, to test for a possible direct effect they are also

included as control variables. Furthermore, this study also includes age and economic situation as

control variables. Age is measured as the number of years since the firm has been incorporated.

Economic situation is measured by making use of a dummy variable. As it is likely that economic

prosperity has an influence on this study, a distinction is made between pre-crisis years (value 0 for

years before 2009) and the years since the beginning of the crisis (value 1 for years from 2009

onward). Including these control variables improves the certainty that the dependent variable is caused

by the independent variable and moderated by the two moderating variables.

29

3.4 Estimation strategy

The approach taken to analyse the data is statistical regression. As the time horizon of this study is

longitudinal, the first step is to identify the most appropriate model. To do so, a couple of tests are

performed. First, a Hausman test is applied. This test indicates whether a fixed effects model or a

random effects model is preferred to analyse the data. Pooled OLS regression is not likely to be an

option, as it does not distinguish between the various firms in the sample; it does not take into account

the possible heterogeneity and individuality that may exist among the firms. However, for certainty a

Breusch-Pagan LM test is performed. This test identifies whether a random effects model or pool OLS

is preferred. Once it has been decided what the most appropriate model is, this model will be applied

in Stata 13 to statistically test the three hypotheses.

Besides the statistical approach, this study will apply another, descriptive, method to

thoroughly analyse the data from a different angle; regardless of whether the statistical methods lead

to conclusive or inconclusive results. This second approach studies more intensively individual firms

or small clusters of firms, possibly displaying and visualizing certain trends or particularities.

3.5 Validity and reliability

3.5.1 Validity

According to Trochim (2006), validity can be defined as “the best available approximation to the

truth of a given proposition, inference, or conclusion”. The overarching term validity includes

construct validity, internal validity and external validity.

External validity is concerned with whether conclusions drawn based on the sample can be

generalized to other settings (Trochim, 2006). Investigating peculiar people, places or times can be a

threat to external validity. However, this study analyses firms from different industries and sizes,

which are spread throughout the Netherlands. Furthermore, based on the facts that longitudinal data

are required and that there is much variation in the date of certification, the threat of conducting a

study in a deviant time is diminished. However, as the sample in this study is not established by using

a random sampling method, the sample does not statistically represent the population. This is likely to

be a considerable threat to external validity.

Internal validity refers to a causal relationship. It addresses the question whether the

independent variable causes the changes in the dependent variable (Trochim, 2006). Criteria are that

the cause happens before the result, that cause and effect covary and that there are no plausible

alternative explanations for the relationship. This is specifically relevant for causal studies. The threat

of reversed causality is somewhat reduced due to the research design. The threat of plausible

alternative explanations is difficult to rule out completely, as this study exclusively contains firms that

30

are certified. However, the inclusion of the moderating and control variables contributes to reducing

this threat.

Construct validity is concerned with really measuring what a study intends to measure. It

refers to whether the operationalization of theoretical constructs accurately represents the theoretical

constructs (Trochim, 2006). The constructs of firm size and industry types are represented accurately

in this study. However, for the independent and dependent variable it can be argued that both

operationalisations do not cover the complete spectrum of the construct. There are many more

measures for firm performance than the one used in this study. Unfortunately, due to time and resource

restrictions, it is not feasible to include numerous measures for firm performance. ROA seems an

appropriate measure, as it is frequently used in similar studies and characterized as a measure of

operating performance. With respect to the independent variable, the literature review identifies that

management systems certification is not the only form of sustainability certification. However,

management system certifications are deeply embedded, require organization-wide changes and are, in

general, well documented. Although it does not cover all aspects of the construct, it is believed to be

an appropriate operationalization.

3.5.2 Reliability

Reliability has to do with repeatability and consistency (Trochim, 2006). It is the extent to which

methods and analysis procedures produce consistent results across occasions and researchers. In

favour of the reliability in this study is the fact that the data used come from public and archival

sources, such as company websites and the databases Orbis and Creditsafe. This is positive for the

repeatability of this study by other researchers. To test for internal consistency, a Cronbach’s Alpha

test is performed in Stata 13.

31

Chapter 4 Data analysis and results

In the current chapter, the data are analysed and the corresponding results are presented. First, an

overview of the data and how they are collected is provided. Then tests are performed to identify the

appropriate model to analyse the data. Subsequently the data are analysed and the hypotheses are

tested, followed by a presentation and short discussion of the results. As mentioned before, this

chapter also includes a descriptive analysis, which is performed after the statistical analysis. Finally,

this chapter ends with some concluding remarks.

4.1 Data collection

Based on the sampling strategy described in the previous chapter, a sample was drawn on the

population. Due to time and resource restrictions, the sample was first refined to the five most

important industries for both certifications. This resulted in a remaining size of 1274 ISO 14001

certifications and 224 OHSAS 18001 certifications. Subsequently, company websites were

investigated to identify the date of initial certification; 447 firms indicated when they obtained their

first certification. The next step was to find firm financial performance and organizational data from

the archival databases for the remaining firms. First, this resulted in the exclusion of firms that got

certified too recently (2014 and 2015), as effects of certification cannot be measured yet. Also firms

that got certified a long time ago are excluded, as most of the required information is not available

from the databases. After checking for the required data, a final sample of 65 firms remained. Table 2

provides an overview of the sample, sorted by industry. The complete sample with elaborate data on

all variables can be found in appendix B.

Table 2: final sample, sorted by industry

SBI Code Industry Dummy Frequency

C. 19-21 Petroleum, chemical and pharmaceutical products D1 4

C. 26-28 Electronics and other machinery/equipment D2 9

E. Waste D3 12

F. Construction D4 21

H. Transport and Storage D5 5

M + N. Business Services D6 14

4.2 The statistical model

As mentioned in chapter 3, before actually analysing the data, it is necessary to test which statistical

model is preferred. First, a Hausman test indicated that the random effects model is preferred over the

32

fixed effects model. The test can be found in appendix C. For robustness a Breusch and Pagan LM test

was performed and indicated that the random effects model is also preferred over pooled OLS

regression. This second test can be found in appendix D. Based on these two tests, this study applies

the random effects model to statistically analyse the data.

4.3 Regression output

Table 3 presents some descriptive statistics of the sample. It can be seen from the table that for some

variables there are no data for all observed years. To test for internal reliability, a Cronbach’s Alpha

test is performed. Unfortunately, the test gives a rather poor value of 0,58. Most likely this is due to

the limited sample size and missing observations for some years. The actual test can be found in

appendix E.

Table 3: descriptive statistics

Variable Obs Mean Std. Dev. Min Max

ROA 329 .0656112 .1107131 -.4817 .6497

Certified 429 .6969697 .4601048 0 1

Age 430 37 28 2 105

Employees 313 217 375 5 2365

Balancesheet* 331 48100 88300 2102,288 569000

Turnover* 200 107000 104000 6992 442000

Year3dum 455 .9054945 .2928527 0 1

D1 455 .0615385 .2405799 0 1

D2 455 .1384615 .3457639 0 1

D3 455 .1846154 .3884123 0 1

D4 455 .3230769 .4681666 0 1

D5 455 .0769231 .2667627 0 1

D6 455 .2153846 .4115414 0 1

* reported in thousands euros

To answer the problem statement, three hypotheses have been developed in chapter 2. As explained

above, the random effects model is used to test them. Below, the models estimated to test the

hypotheses are presented.

H1: Financial performance = 𝛼 + 𝛽1Certified + 𝛽2ln Age + 𝛽3ln Employees + 𝛽4ln Balance sheet +

𝛽5ln Turnover + 𝛽5 Economic situation + 𝛽7D2 + 𝛽8D3 + 𝛽9D4 + 𝛽10D 5+ 𝛽11D6 + ɛ

33

H2: Financial performance = 𝛼 + 𝛽1Certified + 𝛽2ln Age + 𝛽3ln Employees + 𝛽4ln Balance sheet +

𝛽5ln Turnover + 𝛽5 Economic situation + 𝛽7D2 + 𝛽8D3 + 𝛽9D4 + 𝛽10D5 + 𝛽11D6 + 𝛽15Certified*D2 +

𝛽16Certified*D3 + 𝛽17Certified*D4 + 𝛽18Certified*D5 + 𝛽19Certified*D6 + ɛ

H3: Financial performance = 𝛼 + 𝛽1Certified + 𝛽2ln Age + 𝛽3ln Employees + 𝛽4ln Balance sheet +

𝛽5ln Turnover + 𝛽5 Economic situation + 𝛽7D2 + 𝛽8D3 + 𝛽9D4 + 𝛽10D5 + 𝛽11D6 +

𝛽12Certified*Z1Employees + 𝛽13Certified*Z1Balance sheet + 𝛽14Certified*Z1Turnover + ɛ

For hypothesis 1 to be supported, the coefficient of the independent variable (Certified) needs to be

positive and significant. To test for hypotheses 2 and 3, a couple more regressions are run with

interaction terms of the independent and moderating variables being added into the equation. For the

hypotheses to be supported, the coefficients of the interaction terms in each estimated model need to

be significant. Table 4 summarizes the regression results for testing all three hypotheses.

Table 4: Random effects estimations

Model 1 Model 2 Model 3 Model 4 Model 5

Explanatory variables ROA ROA ROA ROA ROA

Certified -0.032* -0.012 -0.033* -0.021

(0.014) (0.027) (0.014) (0.030)

LNAge 0.021t 0.025* 0.026* 0.024* 0.025*

(0.012) (0.012) (0.012) (0.012) (0.012)

LNEmployees -0.006 -0.007 -0.007 -0.006 -0.005

(0.011) (0.010) (0.010) (0.012) (0.012)

LNBalance sheet -0.038* -0.036* -0.034* -0.044* -0.040*

(0.016) (0.016) (0.016) (0.018) (0.018)

LNTurnover 0.015 0.015 0.011 0.019 0.014

(0.020) (0.019) (0.019) (0.020) (0.020)

Economic situation 0.004 -0.005 -0.006 -0.004 -0.005

(0.023) (0.024) (0.024) (0.024) (0.024)

Industry D2 -0.078 -0.078 -0.064 -0.078 -0.074

(0.048) (0.047) (0.054) (0.049) (0.058)

Industry D3 -0.061 -0.062 -0.047 -0.061 -0.052

(0.052) (0.052) (0.063) (0.053) (0.065)

Industry D4 -0.091* -0.094* -0.071 -0.093* -0.078t

(0.040) (0.040) (0.044) (0.041) (0.046)

Industry D5 -0.066 -0.065 -0.084 -0.066 -0.091

34

Model 1 Model 2 Model 3 Model 4 Model 5

Explanatory variables ROA ROA ROA ROA ROA

(0.053) (0.053) (0.062) (0.054) (0.064)

Industry D6 -0.055 -0.055 -0.056 -0.057 -0.058

(0.043) (0.043) (0.041) (0.044) (0.043)

Certified*Industry D2

-0.024 -0.009

(0.044) (0.052)

Certified*Industry D3

-0.023 -0.013

(0.058) (0.061)

Certified*Industry D4

-0.037 -0.027

(0.034) (0.037)

Certified*Industry D5

0.029 0.038

(0.056) (0.058)

o.Certified*Industry D6

- -

Certified*Z1Employees

-0.001 -0.001

(0.010) (0.010)

Certified*Z1Balancesheet

0.010 0.008

(0.010) (0.011)

Certified*Z1turnover

0.002 0.005

(0.034) (0.039)

Constant 0.483* 0.471* 0.479* 0.543* 0.547*

(0.222) (0.220) (0.214) (0.247) (0.252)

Observations 183 183 183 183 183

Number of Firm_id 43 43 43 43 43

Standard errors in parentheses

*** p<0.001, ** p<0.01, * p<0.05, t p<0.10

Model 1 solely includes the control variables. The model indicates that firm age, total balance sheet

and industry dummy 4 have a significant effect on firm financial performance. Whereas, age has a

weak positive effect on ROA, the total balance sheet and operating in the construction industry have a

weak negative effect on ROA. So besides the hypothesized effect in this study, there seem to be other

variables that have a direct effect on firm financial performance.

Model 2 tests for hypothesis 1. As can be seen from this model, certification seems to be significant.

However the sign of the coefficient indicates the opposite direction for the hypothesized relationship;

although a weak coefficient, the model suggests a negative effect of certification on firm performance

35

instead of a positive effect. Therefore, hypothesis 1 is not supported. Comparing this to the literature it

can be stated that this result is not unreasonable. Although many researchers found a positive

relationship, there are also studies that suggest a negative or inconclusive relationship. For model 2 it

is also noteworthy that firm age, total balance sheet and the construction industry, again, seem to have

a significant direct effect on firm performance.

Model 3 is extended with testing for a moderating effect of industry type. As can be identified from

the table, the coefficient of the certification is less negative, but it is not significant anymore as well.

The model also indicates that the interaction terms of the industry dummies with the independent

variable have a weak negative coefficient; only industry dummy 5 has a weak positive coefficient.

However, all of the interaction terms are insignificant, implying that there is no significant moderating

effect of industry type on the relationship between sustainability certification and firm performance.

Therefore, hypothesis 2 is not supported. However, there are still signals for a weak positive

moderating effect. Including industry type as a moderating variable results in the independent variable

becoming less negative and insignificant, which is to some extent, although insignificant, a positive

effect. Looking at the table, this effect might be caused by the interaction term with industry dummy 5

(Transport and storage), which has a positive coefficient. Nevertheless, it is not in line with

expectations to find no significant moderating effect. Previous studies indicate that the effect of

certification on firm performance varies across industries. Also note that the direct effect of industry

dummy 4 is not significant anymore when including industry type as a moderating variable.

Model 4 is test for hypothesis 3, a moderating effect of firm size. In this model, the independent

variable is again significant and negative. The number of employees has a very weak negative

moderating effect, whereas total balance sheet and annual turnover seem to have a very weak positive

moderating effect. However, none of the interaction terms indicate a significant result. Results are

inconclusive; it cannot be stated that firm size has a moderating effect on the relationship between

certification and firm performance and, therefore, hypothesis 3 is not supported. However, as in the

previous models, total balance sheet, which is an aspect of firm size, seems to have a significantly

negative direct effect on firm performance. As for the previous model, finding no moderating effect

for firm size is somewhat surprising as extant literature indicates that smaller firms are likely to

encounter more difficulties in integrating a certified management system.

The final model, model 5, includes all variables. As can be seen from this model, the independent

variable is not significant anymore when including both moderating variables. Moreover, results

indicate again that industry type and firm size are not significant as a moderating variable. However,

as observed from model 3, it is possible that industry type has a very weak positive moderating effect,

making the negative effect of certification insignificant.

36

In conclusion, the final model shows inconclusive results for all three hypotheses. Therefore, none of

the proposed hypotheses is supported. This is not completely surprising, as the sample is rather

limited. However, models 1, 2, 3 and 4 provide some interesting insights. Without including

moderating variables, certification has a significant negative effect on firm performance. The same

result holds when firm size is included as a moderating variable. This suggests the opposite

relationship of what this study proposes. However, including industry type as a moderating variable

makes the independent variable insignificant, pointing to a possible positive moderating effect.

Moreover, some of the control variables seem to have a significant direct effect as well.

To make the results more robust, the sample is tested once again by making use of the fixed

effects model. The fixed effects model also indicates inconclusive results; none of the three

hypotheses is supported in any of the models. The actual results of the fixed effects model can be

found in appendix F.

4.4 Descriptive analysis

As indicated in the previous chapter, this study takes a descriptive approach next to the statistical

analysis. Due to restrictions of the sample, such as small size and missing observations, the statistical

tests lead to inconclusive results. Therefore, the data are analysed once more in a descriptive way.

Analysing the data in this manner reveals some interesting trends and specific features for individual

firms or small clusters of firms. For three firms in the sample, it seems that the year in which the firm

becomes certified is a turning point in these firms’ financial performance. During the two years before

certification, these firms experience decreasing ROAs. However, when these firms become certified,

they start to see an increasing trend with regard to their respective ROAs. Figure 1a depicts the change

in ROA for each of the three firms.

37

Figure 1a: Firms experiencing a downward trend face an increasing ROA after certification

Although no definite conclusions can be drawn from this figure, it is interesting to note that all three

firms are medium sized. Unfortunately, a possible industry effect cannot be identified as the firms

operate in three different industries.

For another group of seven firms in the sample a positive trend can also be recognized. However,

these firms do not experience a particular trend in the years before certification. In the short-term after

certification most of these firms suffer a decrease in ROA, but in the long-run all these firms see their

ROA increasing again. This trend can be seen in figure 1b.

Figure 1b: firms experiencing no particular trend face an increasing ROA after certification

-10,00%

-5,00%

0,00%

5,00%

10,00%

15,00%

20,00%

25,00%

30,00%

35,00%

40,00%

T-2 T-1 T0 T+1 T+2 T+3 T+4

RO

A

Time

From decreasing to increasing ROA

C19-21, medium sized

C26-28, medium sized

F, medium sized

0,00%

5,00%

10,00%

15,00%

20,00%

25,00%

30,00%

T-2 T-1 T0 T+1 T+2 T+3 T+4

RO

A

Time

From no trend to increasing ROA

C26-28, medium sized

C26-28, medium sized

E, medium sized

F, medium sized

H, large

M+N, medium sized

38

As for the previous figure, it is notable to observe the rather similar firm size of the firms displaying

this trend; five medium sized firms and one large firm. Again, with respect to an industry effect,

results are inconclusive as the firms included in the graph operate in five different industries. Two

firms operate in the same industry, but basing a conclusion on this could be misleading.

Whereas figure 1a and 1b depict a positive trend after certification, other firms in the sample show a

significant decreasing trend. For four firms in the sample, it seems that the year of certification is a

turning point in their respective financial performance. However, this time it indicates a negative

turning point. These firms experience increasing ROAs in the years before they become certified, but

after they become certified a trend emerges that shows decreasing ROAs for these four firms. The

change in ROA for these firms is depicted in figure 2a.

Figure 2a: firms experiencing and upward trend face a decreasing ROA after certification

Results are not sufficiently univocal to draw any meaningful conclusions with respect to possible

explanations for this trend. Two of these firms are medium sized and the other two are large firms.

Again, industry effects do not seem to play a role in this case; the four firms operate in four different

industries.

The last peculiarity that is identified in this analysis involves seven firms that display no specific trend

before certification. However, after they obtain the certification, a trend of decreasing ROAs can be

identified for each of these firms. This trend can be seen in figure 2b.

-15,00%

-10,00%

-5,00%

0,00%

5,00%

10,00%

15,00%

20,00%

T-2 T-1 T0 T+1 T+2 T+3 T+4

RO

A

Time

From increasing to decreasing ROA

C26-28, medium sized

F, large

F, medium sized

M+N, large

39

Figure 2b: firms experiencing no particular trend face a decreasing ROA after certification

As mentioned before, it is not appropriate to draw any definite conclusions based on these figures.

However, figure 2b indicates that the firms displaying this particular trend are almost all medium

sized; six out of seven are medium sized, one is a large firm. In this case, industries are also more

concentrated. Three firms operate in the construction sector and another three firms are positioned in

the business services sector. This is not to say that firms in these industries are likely to experience this

trend, but it is a noteworthy observation.

For the remaining firms in the sample, no ongoing positive or negative trend was identified. To some

extent this is in line with the statistical approach; the statistical models show weak coefficients that

are, in most cases, not significant.

Definite conclusions should not be drawn from this descriptive analysis. Although most of the

graphs show that specific trends can be recognized for medium sized firms, it has to be noted that firm

size is not equally distributed across the sample; medium sized firms are overly represented in the

sample. Moreover, these firms are indicated in all of the described trends, so there is not one specific

trend that holds for medium sized firms. This section merely showed that some individual firms

experience a certain trend other than the statistical results show. A tentative conclusion that can be

drawn, however, is that there is not a single relationship that holds for all firms. Unfortunately, from

the data in this sample it is not possible to indicate whether changes are due to firm size or industry

type.

-40,00%

-30,00%

-20,00%

-10,00%

0,00%

10,00%

20,00%

30,00%

T-2 T-1 T0 T+1 T+2 T+3 T+4

RO

A

Time

From no trend to decreasing ROA

C26-28, medium sized

F, medium sized

F, medium sized

F, medium sized

M+N, medium sized

M+N, medium sized

M+N, large

40

4.5 Concluding remarks

This chapter has analysed the data in two ways. First a statistical approach was applied. Testing a

number of models, the results indicated that none of the three hypotheses are supported. Results for the

moderating effects turned out to be inconclusive. In model 2 and 4, the independent variable was

significant, but the coefficient indicated a negative sign, which was the opposite of the hypothesis of

this study. Subsequently, a descriptive analysis was performed. This approach did not lead to any

definite conclusions, but merely identified some interesting trends for individual firms, which deviated

from the statistical results.

Next chapter will discuss the results in more detail, as well as point out limitations of this

study and provide suggestions for future research.

41

Chapter 5 Discussion

In the previous chapter, the data were analysed and the results presented. The present chapter will list

the key findings of this study and, based on that, returns to the empirical research questions that were

formulated in the beginning of this thesis. After ‘answering’ the research questions, this chapter

continuous with comparing these answers to the literature review in chapter 2. Finally, chapter 5 ends

with limitations of the current study and possible avenues for future research.

5.1 Conclusions

In this study the effect of the sustainable management certifications ISO 14001 and OHSAS 18001 on

firm financial performance, in terms of ROA, of Dutch firms was investigated. Moreover, the study

was expanded with assessing possible moderating effects of firm size and industry type on the main

relationship. When looking at the results of the statistical analysis, some interesting findings can be

identified. With respect to the main relationship investigated in this study, the statistical analyses show

either inconclusive results or results that indicate the opposite of what was hypothesized in this study.

It was found that certification, although weakly and not statistically significant for all models, seems

to have a negative effect on firm financial performance. The descriptive analysis nevertheless showed

that some firms deviate from this result and clearly experience an increasing ROA after certification.

However, this holds for a few individual firms, which makes drawing a definite conclusion

inappropriate. Moreover, the same analysis also showed firms experiencing a stronger negative trend

after certification than the statistical results suggest. In conclusion, as certification has a very weak

negative effect, which is not significant for all models, it can be stated that there is no strong evidence

for the effect of sustainability certification on financial performance for Dutch firms. Although some

individual firms show significant changes in financial performance, there is no evidence of a univocal

relationship.

Moving on to the moderating variable firm size, no significant effect was found. This holds for

both the number of employees, total balance sheet and annual turnover. None of these variables has a

significant coefficient, implying that the effect of certification on financial performance does not

become stronger or weaker as a result of firm size. Note, however, that total balance sheet, an aspect

of firm size, seems to have a significant direct negative effect on financial performance. When looking

at the descriptive analysis, there does not appear to be a pattern of firm size either. Medium sized firms

appeared often in the results, but also across different trends, thus observing any specific trends for

firms of a certain size was not possible. Therefore, the conclusion can be drawn that firm size has no

moderating effect on the effect of sustainability certification on financial performance for Dutch firms.

The second moderating variable included in this study was industry type. Six dummies

representing six industries were interacted with certification to identify a possible moderating effect of

42

industry type on the relationship between sustainability certification and financial performance.

However, as for the previous moderating variable, no significant effect was found. None of the six

industries indicates a significant coefficient, implying that the industry in which a firm operates does

not strengthen or weaken the effect of sustainability certification on financial performance. However,

as was found in chapter 4, there are signs for a possible weak moderating effect. The interaction terms

with industry types resulted in the independent variable becoming insignificant and less negative. To

some extent, it can be stated that this is a positive effect. However, effects are still insignificant.

Moreover, the descriptive analysis shows no particular pattern of industries and specific trends.

Consequently, the conclusion can be drawn that industry type does not have a moderating effect on the

relationship between sustainability certification and financial performance for Dutch firms.

5.2 Discussion

With the above conclusions in mind, this section returns to the literature review and compares the

conclusions of this study with the findings of previous research. Although this study has proposed a

positive relationship between sustainability certification and financial performance, finding no support

for this hypothesis was no complete surprise. Previous empirical research had, by no means, led to

conclusive results thus far. As mentioned before, many researchers have found a positive effect of

sustainability and, specifically, sustainability certification on firm performance. Studies indicated that

ISO 14001 certification is likely to improved financial performance, especially in terms of cost

savings (Morrow & Rondinelli, 2002; Lo et al., 2012). Similar results have been found in empirical

studies on the OHSAS 18001 certification. Implementing OHSAS 18001 can lead to increased

benefits and profitability (Fernández-Muñiz et al., 2009; Lo et al., 2014). However, other researchers

have argued that the opposite relationship holds; results indicate that sustainability certification

negatively influences financial performance (Cañón-de-Francia and Garcés-Ayerbe, 2009). As Wang

and Bansal (2012) argued, besides possible positive effects, negative effects such as increased costs

and pressure on management-time are also likely. A third group of research suggests that results are

inconclusive (Heras-Saizarbitoria et al., 2011; Telle, 2006). Sustainability is not likely to lead to any

abnormal changes in financial performance.

Therefore, in line of this extant literature, the results found in this study with respect to the

main relationship are not completely illogical. Although a positive relationship was proposed, the final

model indicated an insignificant relationship, whereas some sub models suggested a weak negative

relationship. As Schaltegger and Synnestvedt (2002) already argued, there is no natural law that

proposes the relationship between sustainability and financial performance.

However, to get a more comprehensive view of the topic, this study included firm size and industry

type to investigate if these variables could possibly strengthen or weaken the effect of sustainability

43

certification on financial performance. Results suggested that none of these variables has a moderating

significant moderating effect. Comparing this to the literature, it is surprising not to find any

significant effects, as previous studies clearly address these aspects and find significant results.

Scholars have suggested that firms operating in varying industries are not likely to experience the

same effect of certification on financial performance. According to the literature, sustainability

certification is, in general, predominantly important for firms in more polluting industries such as

manufacturing and construction, than it is for firms in service industries (Vinodkumar & Bhasi, 2011;

Lo et al., 2014). This is, however, not reflected in the current study, which only finds insignificant

moderating effects for any of the six industries included in the analysis.

Although not many studies distinguish between firm size in their research, scholars argue that

the effects of sustainability certification on financial performance are likely to vary with firm size.

With these sustainability standards being comprehensive systems, literature suggests that larger firms

more often implement and benefit from such certifications (Tinsley, 2002; Fernández-Muñiz et al.,

2009). Integrating sustainability certifications in the core of a business requires many skills and

(financial) resources; resources that smaller firms often lack. Therefore, a moderating effect of firm

size on the relationship between sustainability certification and financial performance was proposed.

However, as for the previous moderating variable, this was not reflected in the results, which indicated

no significant moderating effect of firm size.

In conclusion, it can be stated that the results found in this study are not fully in line with extant

literature. Although previous research has drawn mixed conclusions with respect to the main

relationship in this study, finding inconclusive results with respect to the moderating effects was more

surprising. This might be due to a number of limitations that are discussed below.

5.3 Limitations and future research

The conclusions drawn above have to interpreted with caution, as this study is subject to a number of

limitations. A main and considerable limitation of this study is the restricted sample. Although this

study started very promising with a population of more than 2000 firms, the final sample turned out to

be very small. Unfortunately, the SCCM database with certified firms does not match well with the

financial databases Orbis and Creditsafe. Even for the final 65 firms included in the sample, for some

years observations were missing. This probably has had an impact on the internal reliability as well.

The Cronbach’s Alpha test resulted in a poor value of 0,58, which is below the generally accepted

lower bound 0,7. Moreover, an aspect that might bias the results as well is the unequal distribution of

firm sizes and industry types; medium sized firms and firms from the construction industry are overly

represented. Therefore, to improve the sample size, diminish the number of missing observations and

44

improve the representation of varying firms, future research could apply another strategy, such as

conducting surveys among certified firms.

With respect to statistical analyses this study is limited by the fact that not all models indicate

a significant effect of certification. A possibility is that this has to do with the type of moderating

variable included. The moderating variable industry type, which is a categorical variable, seems to

have an effect on the independent variable. Models without moderators and with firm size as

moderator, a continuous variable, the results show a significant effect for the independent variable.

Models including industry type as a moderator, suggest an insignificant effect. However, due to the

author’s inadequate knowledge of statistical methods, the mechanism of why these types of variables

lead to different results is not clear.

Another limitation of this study is that ROA was used as the only measure for firm

performance. Although ROA is an appropriate and frequently used measure in similar studies, it does

not cover the complete spectrum of firm financial performance. Alternative ratios or different types of

measures might provide other insights, or the explanatory variables might have a different effect on

alternative measures. Thus, future studies could include more than one (type of) measure of firm

financial performance in their research design.

Also, this study might suffer from endogeneity, as all firms in the sample are certified firms.

There could be another variable that has an influence on both certification and firm financial

performance. Therefore, future research could include non-certified firms in the sample as well.

Finally, future research could extend the current study by including aspects of Dutch law and

regulations as variables in the research. Such aspects might provide more elaborate insights in the

relationship proposed in this study.

5.4 Concluding remarks

This section has drawn conclusions based on the results presented in chapter 4. This study has found

no strong evidence for a relationship between sustainability certification and financial performance for

Dutch firms. Moreover, it was concluded that both firm size and industry type have no significant

moderating effect on the main relationship.

Thereafter, the conclusions of the current study were compared to extant literature. This

comparison suggested that the results of the present research were relatively different from what

previous studies have found. However, this could be due to a number of limitations that were

discussed subsequently. These limitations pointed to careful interpretation of the results and

conclusions of this study and they also provided suggestions for future research.

45

References

Ameer, R., & Othman, R. (2012). Sustainability practices and corporate financial performance: A

study based on the top global corporations. Journal of Business Ethics, 108, 61-79.

Balzarova, M., & Castka, P. (2008). Underlying mechanisms in the maintenance of ISO 14001

environmental management system. Journal of Cleaner Production, 16, 1949-1957.

Bansal, P. (2005). Evolving sustainably: A longitudinal study of corporate sustainable development.

Strategic Management Journal, 26, 197–218.

Bansal, P., & Hunter, T. (2003). Strategic Explanations for the Early Adoption of ISO 14001. Journal

of Business Ethics, 46, 289–299.

Beckmann, M., Hielscher, S., & Pies, I. (2014). Commitment strategies for sustainability: How

business firms can transform trade-offs into win-win outcomes. Business Strategy and the

Environment, 23, 18–37.

Bocken, N.M.P., Short, S.W., Rana, P. & Evans, S. (2104). A literature and practice review to develop

sustainable business model archetypes. Journal of Cleaner Production, 65, 42-56.

Bolis, I., Morioka, S., & Sznelwar, L. (2014). When sustainable development risks losing its meaning.

Delimiting the concept with a comprehensive literature review and a conceptual model. Journal of

Cleaner Production, 83, 7-20.

Borga, F., Citterio, A., Noci, G., & Pizzurno, E. (2006). Sustainability report in small enterprises: Case

studies in Italian furniture companies. Business Strategy and the Environment, 18(3), 162-176.

Braungart, M., McDonough, W., & Bollinger, A. (2007) Cradle-to-cradle design: creating healthy

emissions - a strategy for eco-effective product and system design. Journal of Cleaner Production,

15(13-14), 1337-1348.

Brundtland, GH. (1987). Our common future. Oxford: Oxford University Press.

Brunsson, N., Rasche, A., & Seidl, D. (2012) The dynamics of standardization: three perspectives on

standards in organization studies. Organization Studies, 33(5-6), 613-632.

Cañón-de-Francia, J., & Garcés-Ayerbe, C. (2009). ISO 14001 environmental certification: a sign

valued by the market? Environmental and Resource Economics, 44, 245-262.

Carrillo-Hermosilla, J., Del Río, P., & Könnölä, T. (2010). Diversity of eco-innovations: Reflections

from selected case studies. Journal of Cleaner Production, 18, 1073–1083.

Chang, J., & Liang, C. (2009). Performance evaluation of process safety management systems of paint

manufacturing facilities. Journal of Loss Prevention in the Process Industries, 22, 398-402.

46

Cradle to Cradle, get certified. (n.d.). Retrieved April 1, 2015, from http://www.c2ccertified.org/get-

certified/product-certification-process.

Cradle to Cradle, about the institute. (n.d.). Retrieved April 1, 2015, from

http://www.c2ccertified.org/about.

Darnall, N. (2006). Why firms mandate ISO 14001 certification. Business & Society, 45(3), 354–381.

Dovers, SR. (1995). A framework for scaling and framing policy problems in sustainability.

Ecological Economics, 12, 93-106.

Dovers, SR. & Handmer, JW. (1992). Uncertainty, sustainability and change. Global Environmental

Change, 2(4), 262–276.

Elliott, J. (2006). An introduction to sustainable development (3rd ed., pp. 7-43). London: Routledge.

Fairtrade, history of. (n.d.). Retrieved March 31, 2015, from http://www.fairtrade.net/history-of-

fairtrade.html.

Fairtrade, certifying. (n.d.). Retrieved March 31, 2015, from http://www.fairtrade.net/certifying-

fairtrade.html.

Fernández-Muñiz, B., Montes-Peón, J., & Vázquez-Ordás, C. (2009). Relation Between Occupational

Safety Management And Firm Performance. Safety Science, 47, 980-991.

Fernández-Muñiz, B., Montes-Peón, J., & Vázquez-Ordás, C. (2012). Occupational risk management

under the OHSAS 18001 standard: Analysis of perceptions and attitudes of certified firms. Journal of

Cleaner Production, 24, 36-47.

Gimenez, C., Sierra, V. & Rodon, J. (2012). Sustainable operations: their impact on the triple bottom

line. International Journal of Production Economics, 140, 149-159.

Giovannucci, D., & Ponte, S. (2005) Standards as a new form of social contract? Sustainability

initiatives in the coffee industry. Food Policy, 30, 284-301.

Gladwin, T., Kennelly, J., & Krause, T. (1995). Shifting paradigms for sustainable development:

Implications for management theory and research. Academy of Management Review, 20(4), 874-907.

Hall, JK., Daneke, GA. & Lenox, MJ. (2010). Sustainable development and entrepreneurship: past

contributions and future directions. Journal of Business Venturing, 25, 439-448.

Hart, S.L., & Ahuja, G. (1996). Does it pay to be green? An empirical examination of the relationship

between emission reduction and firm performance. Business Strategy and the Environment, 5, 30-37.

Henriques, A. & Richardson, J. (2004). The triple bottom line: Does it all add up?. (1st ed.). London:

Earthscan.

47

Heras, I., & Arana, G. (2010). Alternative models for environmental management in SMEs: The case

of Ekoscan vs. ISO 14001. Journal of Cleaner Production, 726-735.

Heras-Saizarbitoria, I., & Boiral, O. (2013). ISO 9001 and ISO 14001: Towards a research agenda on

management system standards. International Journal of Management Reviews, 15(1), 47-65.

Holden, E., Linnerud, K. & Banister, D. (2014) Sustainable development: Our Common Future

revisited. Global Environmental Change, 26, 130-139.

Hubbard, G. (2009). Measuring organizational performance: beyond the triple bottom line. Business

Strategy and the Environment, 19, 177-191.

ISO, ISO 1400 - Environmental Management. (n.d.). Retrieved April 1, 2015, from

http://www.iso.org/iso/home/standards/management-standards/iso14000.html.

Jørgensen, T. (2008). Towards more sustainable management systems: Through life cycle

management and integration. Journal of Cleaner Production, 16, 1071-1080.

Jørgensen, T., Remmen, A., & Mellado, M. (2006). Integrated management systems – three different

levels of integration. Journal of Cleaner Production, 14, 713-722.

King, A.A., Lenox, M.J., & Terlaak, A. (2005). The strategic use of decentralized institutions:

exploring certification with the ISO 14001 management standard. The Academy of Management

Journal, 48(6), 1091-1106.

Klewitz, J., & Hansen, E. (2014). Sustainability-oriented innovation of SMEs: A systematic review.

Journal of Cleaner Production, 65.

Lo, C.K.Y, Pagell, M., Fan, D., Wiengarten, F., & Yeung, A.C.L. (2014). OHSAS 18001 certification

and operating performance: The role of complexity and coupling. Journal of Operations Management,

32, 268-280.

López, M.V., Garcia, A., & Rodriguez, L. (2007). Sustainable development and corporate

performance: A study based on the Dow Jones sustainability index. Journal of Business Ethics, 75,

285–300.

Mawhinney, M. (2002). Sustainable development: Understanding the green debates (pp. 1-10).

Malden, MA: Blackwell Science.

McDonough, W., & Braungart, M. (2003). Towards a sustaining architecture for the 21st century: The

promise of cradle-to-cradle design. Industry & Environment, 26(2-3), 13-16.

Miles, M.P., & Covin, J.G. (2000). Environmental marketing: A source of reputational, competitive

and financial advantage. Journal of Business Ethics, 23, 299-311.

Miles, M., Munilla, L., & Darroch, J. (2009). Sustainable corporate entrepreneurship. International

entrepreneurship and management journal, 5, 65-76.

48

Morrow, D., & Rondinelli, D. (2002). Adopting corporate environmental management systems:

motivations and results of ISO 14001 and EMAS certification. European Management Journal, 20(2),

159-171.

Nidumolu, R., Prahalad, C.K. and Rangaswami, M.R. (2009) Why sustainability is now the key driver

of innovation. Harvard Business Review, 1-10.

Norman, W., & MacDonald, C. (2004). Getting to the bottom of “triple bottom line”. Business Ethics

Quarterly, 14(2), 243-262.

O’Dwyer, B., & Owen, D.L. (2005). Assurance statement practice in environmental, social and

sustainability reporting: a critical evaluation. The British Accounting Review, 37(2), 205-229.

Parrish, B.D. (2010). Sustainability-driven entrepreneurship: Principles of organization design.

Journal of Business Venturing, 25(5), 510–523.

Plaza-Úbeda, J.A., Burgos-Jiménez, J., Vazquez, D.A., & Liston-Heyes, C. (2009). The win-win

paradigm and stakeholder integration. Business Strategy and the Environment, 18, 487–499.

Porter, M.E., & Kramer, M.R. (2011). Creating shared value. Harvard Business Review.

Porter, M.E., & Reindhardt, F.L. (2007). A strategic approach to climate. Harvard Business Review,

85(10), 22-25.

Potts, J., Lynch, M., Wilkings, A., Huppé, G., Cunningham, M., & Voora, V. (2014). The state of

sustainability initiatives review 2014. International Institute for Sustainable Development. Retrieved

March 31, 2015, from https://www.iisd.org/pdf/2014/ssi_2014.pdf.

Rainforest Alliance, certification, verification and validation services. (n.d.). Retrieved March 31,

2015, from http://www.rainforest-alliance.org/certification-verification.

Ramus, C.A., & Montiel, I. (2005). When are corporate environmental policies a form of

greenwashing? Business & Society, 44(4), 377–414.

Reinecke, J., Manning, S., & Von Hagen, O. (2012). The emergence of a standards market:

multiplicity of sustainability standards in the global coffee industry. Organization Studies, 33(5-6),

791-814.

Roberts, D.H., & Koeplin, J.P. (2007). Sustainability reporting practices in Portugal: greenwashing or

triple bottom line? International Business and Economics Research Journal, 6(9), 29-40.

Santos, G., Mendes, F., & Barbosa, J. (2011). Certification and integration of management systems:

the experience of Portuguese small and medium enterprises. Journal of Cleaner Production, 19, 1965-

1974.

49

Schaltegger, S., & Synnestvedt, T. (2002). The link between “green” and economic success:

environmental management as the crucial trigger between environmental and economic performance.

Journal of Environmental Management, 65, 339–346.

Steurer, R. (2006). Mapping stakeholder theory anew: from the ‘stakeholder theory of the firm’ to

three perspectives on business–society relations. Business Strategy and the Environment, 15, 55–69.

Telle, K. (2006). “It pays to be green” – a premature conclusion? Environmental and Resource

Economics, 35, 195-220.

Tinsley, S. (2002). EMS models for business strategy development. Business Strategy and the

Environment, 11, 376-390.

Trochim, W.M. (2006) The Research Methods Knowledge Base, 2nd Edition. Retrieved, 15 April,

from http://www.socialresearchmethods.net/kb/.

UTZ, What is UTZ certified. (n.d.). Retrieved March 31, 2015, from

https://www.utzcertified.org/aboututzcertified/whatisutzcertified.

Van Marrewijk, M. (2003). Concepts and definitions of csr and corporate responsibility: Between

agency and communion. Journal of business ethics, 44, 95-105.

Vinodkumar, M., & Bhasi, M. (2011). A study on the impact of management system certification on

safety management. Safety Science, 49, 498-507.

Von Hagen, O., Manning, S., & Reinecke, J. (2010). Sustainable sourcing in the food industry: global

challenges and practices. Moderne Ernährung Heute, 4, 1-9.

Wang, T., & Bansal, P. (2012). Social responsibility in new ventures: profiting from a long-term

orientation. Strategic Management Journal, 33, 1135-1153.

White, M.A. (2013) Sustainability: I know it when I see it. Ecological Economics, 86, 213-217.

Yap, N. (1989-90). NGOs and sustainable development. International Journal, XLV Winter, 75-105.

50

Appendices

Appendix A – Certifications sorted by industry

OHSAS 18001 Certification

Industry Frequency %

A. Landbouw, bosbouw, visserij 8 1,80%

B. Winning van delfstoffen 20 4,49%

C. 10-12 Voedings- en genotsmiddelen 14 3,15%

C. 13-15 Textiel, kleding, leer 3 0,67%

C. 16 Primaire houtbewerking 1 0,22%

C. 17 Papier en karton 19 4,27%

C. 18 Drukkerijen 2 0,45%

C. 19-21 Aardolie, chemische en farmaceutische producten 32 7,19%

C. 22 Rubber en kunststof 8 1,80%

C. 23 Overige niet-metaalhoudende mineralen 13 2,92%

C. 24 Primaire metaalbewerking 4 0,90%

C. 25 Metaal producten (geen machines) 28 6,29%

C. 26-28 Elektronica en overige machines/apparatuur 46 10,34%

C. 29-30 Auto's en transportmiddelen 10 2,25%

C. 31 Meubels 0 0,00%

C. 32 Overige goederen 4 0,90%

C. 33 Reparatie en installatie van apparaten en machines 12 2,70%

D. Elektriciteit, aardgas, stoom, gekoelde lucht 12 2,70%

E. Afval 27 6,07%

F. Bouwnijverheid 42 9,44%

G. Groot- en detailhandel 13 2,92%

H. Vervoer en opslag 48 10,79%

I. Logies-, maaltijd- en drankverstrekking 1 0,22%

J. Informatie en communicatie 6 1,35%

K. Financiële instellingen 1 0,22%

L. Onroerend goed 0 0,00%

M + N. Zakelijke dienstverlening 56 12,58%

O + P. Overheid en onderwijs 14 3,15%

Q. Gezondheidszorg 1 0,22%

R. Cultuur, sport en recreatie 0 0,00%

Total 445 100,00%

51

ISO 14001 Certification

Industry Frequency %

A. Landbouw, bosbouw, visserij 30 1,25%

B. Winning van delfstoffen 36 1,51%

C. 10-12 Voedings- en genotsmiddelen 74 3,09%

C. 13-15 Textiel, kleding, leer 20 0,84%

C. 16 Primaire houtbewerking 8 0,33%

C. 17 Papier en karton 33 1,38%

C. 18 Drukkerijen 51 2,13%

C. 19-21 Aardolie, chemische en farmaceutische producten 126 5,27%

C. 22 Rubber en kunststof 91 3,80%

C. 23 Overige niet-metaalhoudende mineralen 58 2,42%

C. 24 Primaire metaalbewerking 23 0,96%

C. 25 Metaal producten (geen machines) 169 7,07%

C. 26-28 Elektronica en overige machines/apparatuur 178 7,44%

C. 29-30 Auto's en transportmiddelen 22 0,92%

C. 31 Meubels 23 0,96%

C. 32 Overige goederen 15 0,63%

C. 33 Reparatie en installatie van apparaten en machines 52 2,17%

D. Elektriciteit, aardgas, stoom, gekoelde lucht 17 0,71%

E. Afval 196 8,19%

F. Bouwnijverheid 363 15,18%

G. Groot- en detailhandel 163 6,81%

H. Vervoer en opslag 187 7,82%

I. Logies-, maaltijd- en drankverstrekking 12 0,50%

J. Informatie en communicatie 51 2,13%

K. Financiële instellingen 2 0,08%

L. Onroerend goed 7 0,29%

M + N. Zakelijke dienstverlening 350 14,63%

O + P. Overheid en onderwijs 27 1,13%

Q. Gezondheidszorg 6 0,25%

R. Cultuur, sport en recreatie 2 0,08%

Total 2392 100,00%

52

Appendix B – Sample with data

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t-2 1 7,43% 0 63 128 9511679,96 0 0 1 0 0 0

t-1 1 6,10% 0 64 128 9211543,00 0 0 1 0 0 0

t0 1 7,59% 1 65 127 9045117,00 0 0 1 0 0 0

t+1 1 1,93% 1 66 127 10118773,00 0 0 1 0 0 0

t+2 1 0,47% 1 67 122 9152895,00 0 0 1 0 0 0

t+3 1 19,20% 1 68 122 8714987,00 0 0 1 0 0 0

t+4 1

1 69

0 0 1 0 0 0

t-2 2 3,44% 0 91 62 4688286,00 0 0 0 1 0 0

t-1 2 9,83% 0 92 59 3943266,00 0 0 0 1 0 0

t0 2 -48,17% 1 93 48 2500170,00 0 0 0 1 0 0

t+1 2 -46,20% 1 94 37 2287198,00 0 0 0 1 0 0

t+2 2

1 95

0 0 0 1 0 0

t+3 2

1 96

0 0 0 1 0 0

t+4 2 0 0 0 1 0 0

t-2 3

0 21

0 0 0 1 0 0

t-1 3 4,93% 0 22 185 34981000,00 55397000,00 0 0 0 1 0 0

t0 3 4,60% 1 23 185 37647000,00 58093000,00 0 0 0 1 0 0

t+1 3 3,46% 1 24 185 34728000,00 57229000,00 0 0 0 1 0 0

t+2 3 3,66% 1 25 42079000,00 60248000,00 0 0 0 1 0 0

t+3 3 7,21% 1 26 40634000,00 59958000,00 0 0 0 1 0 0

t+4 3

1 27

0 0 0 1 0 0

t-2 4 0 25 0 1 0 0 0 0

t-1 4

0 26

0 1 0 0 0 0

t0 4 2,42% 1 27 190 211981000,00 316270000,00 0 1 0 0 0 0

t+1 4 5,03% 1 28 190

0 1 0 0 0 0

53

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+2 4

1 29

0 1 0 0 0 0

t+3 4

1 30

0 1 0 0 0 0

t+4 4 1 31 0 1 0 0 0 0

t-2 5 -0,55% 0 12 81 7684845,87 19397858,73 0 1 0 0 0 0

t-1 5 3,33% 0 13 68 10042000,00 16975000,00 0 1 0 0 0 0

t0 5 6,66% 1 14 63 7984000,00 14206000,00 0 1 0 0 0 0

t+1 5 11,72% 1 15 60 8476000,00 14557000,00 0 1 0 0 0 0

t+2 5 10,70% 1 16 61 6203000,00 13651000,00 0 1 0 0 0 0

t+3 5 -12,24% 1 17 88 7195189,00 12921306,00 0 1 0 0 0 0

t+4 5

1 18

0 1 0 0 0 0

t-2 6 0,26% 0 50 20031520,00 34576780,00 0 0 0 1 0 0

t-1 6 1,29% 0 51 122 22911020,00 43951411,00 0 0 0 1 0 0

t0 6 4,12% 1 52 28106512,00 41008907,00 0 0 0 1 0 0

t+1 6 -0,65% 1 53 135 26723494,00 55234767,00 0 0 0 1 0 0

t+2 6

1 54

0 0 0 1 0 0

t+3 6

1 55

0 0 0 1 0 0

t+4 6 0 0 0 1 0 0

t-2 7 -0,36% 0 70 71 2658510,00 0 0 0 0 0 1

t-1 7 -5,93% 0 71 74 2880371,00 0 0 0 0 0 1

t0 7 -0,25% 1 72 84 3462421,00 0 0 0 0 0 1

t+1 7 -8,67% 1 73 86 4589954,00 0 0 0 0 0 1

t+2 7 6,70% 1 74 76 5574686,00 0 0 0 0 0 1

t+3 7

1 75

0 0 0 0 0 1

t+4 7

1 76

0 0 0 0 0 1

t-2 8 7,94% 0 22 7 58572688,00 52565899,00 0 0 0 1 0 0

t-1 8 14,67% 0 23 157 83744691,00 81993782,00 0 0 0 1 0 0

t0 8 10,04% 1 24 156 78216355,00 66512274,00 0 0 0 1 0 0

t+1 8 3,82% 1 25 165 80228672,00 59257808,00 0 0 0 1 0 0

t+2 8

1 26

0 0 0 1 0 0

54

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+3 8

1 27

0 0 0 1 0 0

t+4 8 0 0 0 1 0 0

t-2 9 -15,78% 0 10 117 13763151,00 14904422,92 0 0 0 0 0 1

t-1 9 37,76% 0 11 123 13038176,00 22803356,00 0 0 0 0 0 1

t0 9 64,97% 1 12 115 6899951,00 20182686,00 0 0 0 0 0 1

t+1 9 -22,07% 1 13 120 6791420,00 17110978,00 0 0 0 0 0 1

t+2 9

1 14

0 0 0 0 0 1

t+3 9

1 15

0 0 0 0 0 1

t+4 9

0 0 0 0 0 1

t-2 10 24,55% 0 85 151 18247781,00 60265653,00 0 0 0 1 0 0

t-1 10 18,85% 0 86 151 20332888,00 56638608,00 0 0 0 1 0 0

t0 10 10,37% 1 87 150 22149490,00 45872115,00 0 0 0 1 0 0

t+1 10 6,67% 1 88 145 23884710,00 45096824,00 0 0 0 1 0 0

t+2 10 9,29% 1 89 145 21671634,00 48126382,00 0 0 0 1 0 0

t+3 10

1 90

0 0 0 1 0 0

t+4 10 1 91 0 0 0 1 0 0

t-2 11 7,35% 0 13 13 20307643,00 93539840,00 0 0 0 0 0 1

t-1 11 3,69% 0 14 17 27378627,00 110626742,00 0 0 0 0 0 1

t0 11 4,56% 1 15 21 40910734,00 183254993,00 0 0 0 0 0 1

t+1 11 4,33% 1 16 25 48304295,00 241480692,00 0 0 0 0 0 1

t+2 11 3,44% 1 17 26 51003731,00 262843695,00 0 0 0 0 0 1

t+3 11

1 18

0 0 0 0 0 1

t+4 11

1 19

0 0 0 0 0 1

t-2 12 13,67% 0 46 250 136396000,00 325196000,00 1 0 0 0 0 0

t-1 12 15,29% 0 47 252 163186000,00 365248000,00 1 0 0 0 0 0

t0 12 13,71% 1 48 255 192778000,00 380833000,00 1 0 0 0 0 0

t+1 12 12,64% 1 49 252 212292000,00 360973000,00 1 0 0 0 0 0

t+2 12

1 50

1 0 0 0 0 0

t+3 12

1 51

1 0 0 0 0 0

55

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+4 12 1 0 0 0 0 0

t-2 13 1,44% 0 16 6 33562601,00 0 0 1 0 0 0

t-1 13 -0,04% 0 17 6 32567408,00 0 0 1 0 0 0

t0 13 2,52% 1 18 6 31607441,00 0 0 1 0 0 0

t+1 13 -3,13% 1 19 102 25731116,00 0 0 1 0 0 0

t+2 13

1 20

0 0 1 0 0 0

t+3 13

1 21

0 0 1 0 0 0

t+4 13

0 0 1 0 0 0

t-2 14 -3,21% 0 18 206 120547904,70 127942259,10 0 0 0 1 0 0

t-1 14 2,68% 0 19 209 93900457,00 136797872,00 0 0 0 1 0 0

t0 14 3,76% 1 20 5 107855135,00 125855671,00 0 0 0 1 0 0

t+1 14 4,22% 1 21 175 97265664,00 124511946,00 0 0 0 1 0 0

t+2 14 0,93% 1 22 180 106267052,00 145023518,00 0 0 0 1 0 0

t+3 14 0,95% 1 23 175 95788486,00 133571000,00 0 0 0 1 0 0

t+4 14 1 24 0 0 0 1 0 0

t-2 15 4,08% 0 13 5 6767996,70 49897975,14 0 0 1 0 0 0

t-1 15 2,45% 0 14 9 8575650,00 36195199,00 0 0 1 0 0 0

t0 15 3,49% 1 15 9 9788808,00 44749727,00 0 0 1 0 0 0

t+1 15 0,83% 1 16 9160561,00 41792878,00 0 0 1 0 0 0

t+2 15 -9,58% 1 17 7251951,00 27028690,00 0 0 1 0 0 0

t+3 15 1,54% 1 18 10 5098744,00 0 0 1 0 0 0

t+4 15

1 19

0 0 1 0 0 0

t-2 16 15,81% 0 24 86 9965713,16 0 1 0 0 0 0

t-1 16 -6,36% 0 25 92 9006601,00 0 1 0 0 0 0

t0 16 0,73% 1 26 94 9898790,00 0 1 0 0 0 0

t+1 16 -5,79% 1 27 93 10054431,00 0 1 0 0 0 0

t+2 16 -6,96% 1 28 89 8754314,00 0 1 0 0 0 0

t+3 16 -7,96% 1 29 77 9578247,00 0 1 0 0 0 0

t+4 16 1 30 0 1 0 0 0 0

56

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t-2 17 8,54% 0 96 104567076,20 322476000,11 0 0 0 1 0 0

t-1 17 5,92% 0 97 1861 128436762,50 277521000,87 0 0 0 1 0 0

t0 17 5,66% 1 98 1884 107523000,00 266961000,00 0 0 0 1 0 0

t+1 17 6,13% 1 99 1941 112273000,00 327444000,00 0 0 0 1 0 0

t+2 17 5,27% 1 100 1986 120449000,00 386029000,00 0 0 0 1 0 0

t+3 17 2,52% 1 101 2365 163662000,00 441642000,00 0 0 0 1 0 0

t+4 17 3,04% 1 102 2258 146023000,00 391267000,00 0 0 0 1 0 0

t-2 18 2,82% 0 19 25 7529464,00 0 0 0 0 0 1

t-1 18 5,37% 0 20 29 8758553,00 0 0 0 0 0 1

t0 18 -6,54% 1 21 29 8676582,00 0 0 0 0 0 1

t+1 18 2,92% 1 22 39 9493483,00 0 0 0 0 0 1

t+2 18

1 23

0 0 0 0 0 1

t+3 18

1 24

0 0 0 0 0 1

t+4 18 0 0 0 0 0 1

t-2 19 3,05% 0 72 49 10463817,85 0 1 0 0 0 0

t-1 19 6,42% 0 73 51 10781811,00 0 1 0 0 0 0

t0 19 2,59% 1 74 48 11244407,00 0 1 0 0 0 0

t+1 19 5,09% 1 75 45 10568580,00 0 1 0 0 0 0

t+2 19 12,64% 1 76 46 13064718,00 0 1 0 0 0 0

t+3 19 13,02% 1 77 51 16968585,00 0 1 0 0 0 0

t+4 19

1 78

0 1 0 0 0 0

t-2 20 0 2 1 0 0 0 0 0

t-1 20 -2,42% 0 3 125 95385231,00 142343969,00 1 0 0 0 0 0

t0 20 1,08% 1 4 138 111134413,00 167629262,00 1 0 0 0 0 0

t+1 20 4,50% 1 5 142 104790701,00 185235589,00 1 0 0 0 0 0

t+2 20

1 6

1 0 0 0 0 0

t+3 20

1 7

1 0 0 0 0 0

t+4 20 8 1 0 0 0 0 0

t-2 21 8,57% 0 45 367 68535332,50 161257000,81 0 0 0 1 0 0

57

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t-1 21 8,72% 0 46 360 71346000,00 157211000,00 0 0 0 1 0 0

t0 21 8,92% 1 47 355 71417000,00 154231000,00 0 0 0 1 0 0

t+1 21 4,10% 1 48 473 82725000,00 260488000,00 0 0 0 1 0 0

t+2 21 1,56% 1 49 379 70026000,00 196175000,00 0 0 0 1 0 0

t+3 21 0,35% 1 50 280 52059000,00 122152000,00 0 0 0 1 0 0

t+4 21

1 51

0 0 0 1 0 0

t-2 22 12,59% 0 26 49 12875020,48 24299443,00 0 0 0 1 0 0

t-1 22 9,60% 0 27 49 16736203,00 33282805,00 0 0 0 1 0 0

t0 22 20,20% 1 28 51 19420353,00 43166387,00 0 0 0 1 0 0

t+1 22 5,23% 1 29 53 19408191,00 43026316,00 0 0 0 1 0 0

t+2 22 3,48% 1 30 55 19979559,00 31399308,00 0 0 0 1 0 0

t+3 22 -0,12% 1 31 56 25243214,00 34790011,00 0 0 0 1 0 0

t+4 22 1 32 0 0 0 1 0 0

t-2 23

0 5 9787701,00 0 0 1 0 0 0

t-1 23 -0,32% 0 6 41 28025229,00 0 0 1 0 0 0

t0 23 -1,74% 1 7 41 25873829,00 0 0 1 0 0 0

t+1 23 -0,73% 1 8 35 23266941,00 0 0 1 0 0 0

t+2 23 0,49% 1 9 31 22845936,00 0 0 1 0 0 0

t+3 23

1 10

0 0 1 0 0 0

t+4 23

1 11

0 0 1 0 0 0

t-2 24 9,25% 0 33 120 20410457,00 51817471,00 0 0 0 1 0 0

t-1 24 6,44% 0 34 121 21789477,00 46712782,00 0 0 0 1 0 0

t0 24 3,60% 1 35 119 21537138,00 62271615,00 0 0 0 1 0 0

t+1 24 0,51% 1 36 119 17566129,00 40746841,00 0 0 0 1 0 0

t+2 24

1 37

0 0 0 1 0 0

t+3 24

1 38

0 0 0 1 0 0

t+4 24 0 0 0 1 0 0

t-2 25 -3,83% 0 28 213 18117987,00 56870500,00 0 0 0 1 0 0

t-1 25 0,75% 0 29 189 14272287,00 45853138,00 0 0 0 1 0 0

58

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t0 25 1,17% 1 30 147 14259225,00 41303712,00 0 0 0 1 0 0

t+1 25 1,05% 1 31 191 16917302,00 12901855,00 0 0 0 1 0 0

t+2 25

1 32

0 0 0 1 0 0

t+3 25

0 0 0 1 0 0

t+4 25

0 0 0 1 0 0

t-2 26 5,07% 0 16 176 61643661,71 410303161,60 0 0 1 0 0 0

t-1 26 5,20% 0 17 167 78814817,00 221249968,00 0 0 1 0 0 0

t0 26 6,19% 1 18 161 90488285,00 354574967,00 0 0 1 0 0 0

t+1 26 3,85% 1 19 165 86096143,00 396753497,00 0 0 1 0 0 0

t+2 26 -2,11% 1 20 174 90178141,00 347807574,00 0 0 1 0 0 0

t+3 26 -0,50% 1 21 177 75129897,00 304284333,00 0 0 1 0 0 0

t+4 26 1 22 0 0 1 0 0 0

t-2 27 1,61% 0 20 117 28063451,00 0 0 1 0 0 0

t-1 27 2,17% 0 21 126 30370111,00 0 0 1 0 0 0

t0 27 0,51% 1 22 141 33298706,00 0 0 1 0 0 0

t+1 27 -3,80% 1 23 142 36561670,00 0 0 1 0 0 0

t+2 27

1 24

0 0 1 0 0 0

t+3 27

1 25

0 0 1 0 0 0

t+4 27

0 0 1 0 0 0

t-2 28 3,99% 0 59 1652 144462000,00 311336000,00 0 0 0 0 0 1

t-1 28 5,52% 0 60 1630 117325000,00 294102000,00 0 0 0 0 0 1

t0 28 5,25% 1 61 1716 115499000,00 321694000,00 0 0 0 0 0 1

t+1 28 4,86% 1 62 1781 104848000,00 314399000,00 0 0 0 0 0 1

t+2 28 -3,99% 1 63 1856 117858000,00 290253000,00 0 0 0 0 0 1

t+3 28

1 64

0 0 0 0 0 1

t+4 28 1 65 0 0 0 0 0 1

t-2 29 15,07% 0 86 96 9939905,00 38928237,00 0 0 0 1 0 0

t-1 29 18,50% 0 87 96 9680614,00 36845944,00 0 0 0 1 0 0

t0 29 0,45% 1 88 103 7543023,00 51774032,00 0 0 0 1 0 0

59

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+1 29 -6,96% 1 89 88 5065247,00 26621289,00 0 0 0 1 0 0

t+2 29 3,69% 1 90 67 5018584,00 21424026,00 0 0 0 1 0 0

t+3 29

1 91

0 0 0 1 0 0

t+4 29

1 92

0 0 0 1 0 0

t-2 30 4,96% 0 25 147 14143000,00 27926000,00 0 0 0 1 0 0

t-1 30 5,89% 0 26 147 15194000,00 38440000,00 0 0 0 1 0 0

t0 30 4,78% 1 27 157 14991000,00 34186000,00 0 0 0 1 0 0

t+1 30 3,57% 1 28 151 17019000,00 35215000,00 0 0 0 1 0 0

t+2 30 1,14% 1 29 140 15566000,00 29602000,00 0 0 0 1 0 0

t+3 30

1 30

0 0 0 1 0 0

t+4 30 1 31 0 0 0 1 0 0

t-2 31 36,93% 0 72 49 5209773,00 10675453,00 0 1 0 0 0 0

t-1 31 28,61% 0 73 43 5727420,00 9473959,00 0 1 0 0 0 0

t0 31 16,09% 1 74 47 7687137,00 12182506,00 0 1 0 0 0 0

t+1 31 15,86% 1 75 52 6415877,00 13539369,00 0 1 0 0 0 0

t+2 31 30,72% 1 76 52 7364098,00 14552050,00 0 1 0 0 0 0

t+3 31

1 77

0 1 0 0 0 0

t+4 31

1 78

0 1 0 0 0 0

t-2 32 -1,35% 0 5 65 3777488,00 0 0 1 0 0 0

t-1 32 -5,52% 0 6 72 3547899,00 0 0 1 0 0 0

t0 32 0,20% 1 7 71 4412092,00 0 0 1 0 0 0

t+1 32 9,90% 1 8 65 3910825,00 0 0 1 0 0 0

t+2 32 1,24% 1 9 63 3999270,00 0 0 1 0 0 0

t+3 32

1 10

0 0 1 0 0 0

t+4 32 1 11 0 0 1 0 0 0

t-2 33 23,30% 0 2 24 11922109,65 0 0 0 0 0 1

t-1 33 16,26% 0 3 30 24539000,00 0 0 0 0 0 1

t0 33 19,95% 1 4 37 16572000,00 20144000,00 0 0 0 0 0 1

t+1 33 -3,55% 1 5 50 33230000,00 57265000,00 0 0 0 0 0 1

60

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+2 33 -15,68% 1 6 56 41488000,00 77941000,00 0 0 0 0 0 1

t+3 33 -35,16% 1 7 64 37726000,00 27433000,00 0 0 0 0 0 1

t+4 33

1 8

0 0 0 0 0 1

t-2 34 10,34% 0 75 53 8063268,00 0 0 0 1 0 0

t-1 34 8,57% 0 76 57 7831632,00 0 0 0 1 0 0

t0 34 6,67% 1 77 59 10002680,00 0 0 0 1 0 0

t+1 34 14,47% 1 78 64 10257382,00 0 0 0 1 0 0

t+2 34

1 79

0 0 0 1 0 0

t+3 34

1 80

0 0 0 1 0 0

t+4 34 0 0 0 1 0 0

t-2 35 1,40% 0 34 147892000,00 112999000,00 0 0 0 0 0 1

t-1 35 3,52% 0 35 60 173393000,00 132569000,00 0 0 0 0 0 1

t0 35 2,97% 1 36 181965000,00 143383000,00 0 0 0 0 0 1

t+1 35 3,13% 1 37 50 177581000,00 138896000,00 0 0 0 0 0 1

t+2 35

1 38

0 0 0 0 0 1

t+3 35

1 39

0 0 0 0 0 1

t+4 35

0 0 0 0 0 1

t-2 36 5,21% 0 10 541 19156819,00 65703209,00 0 0 0 0 1 0

t-1 36 3,61% 0 11 559 18409465,00 77012243,00 0 0 0 0 1 0

t0 36 6,35% 1 12 558 19022054,00 85989405,00 0 0 0 0 1 0

t+1 36 11,61% 1 13 540 22493710,00 100819029,00 0 0 0 0 1 0

t+2 36 12,11% 1 14 516 25345389,00 112096097,00 0 0 0 0 1 0

t+3 36

1 15

0 0 0 0 1 0

t+4 36 1 16 0 0 0 0 1 0

t-2 37 16,74% 0 14 7158467,00 26649360,00 0 0 0 0 1 0

t-1 37 14,99% 0 15 64 6322079,00 26121423,00 0 0 0 0 1 0

t0 37 11,77% 1 16 64 6617142,00 24159685,00 0 0 0 0 1 0

t+1 37 11,68% 1 17 57 8036775,00 24272882,00 0 0 0 0 1 0

t+2 37

1 18

0 0 0 0 1 0

61

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+3 37

1 19

0 0 0 0 1 0

t+4 37

0 0 0 0 1 0

t-2 38 10,01% 0 4 57667000,00 117420000,00 0 0 0 1 0 0

t-1 38 7,32% 0 5 53552000,00 116142000,00 0 0 0 1 0 0

t0 38 -5,82% 1 6 42196000,00 89492000,00 0 0 0 1 0 0

t+1 38 0,33% 1 7 46223000,00 83892000,00 0 0 0 1 0 0

t+2 38 11,61% 1 8 24926000,00 6992000,00 0 0 0 1 0 0

t+3 38

1 9

0 0 0 1 0 0

t+4 38 1 10 0 0 0 1 0 0

t-2 39 8,56% 0 54 29 23772653,00 0 0 0 0 0 1

t-1 39 5,94% 0 55 31 21900356,00 0 0 0 0 0 1

t0 39 9,02% 1 56 59 25771329,00 0 0 0 0 0 1

t+1 39 6,14% 1 57 114 43994276,00 0 0 0 0 0 1

t+2 39 3,60% 1 58 122 55807935,00 0 0 0 0 0 1

t+3 39

1 59

0 0 0 0 0 1

t+4 39

1 60

0 0 0 0 0 1

t-2 40 -6,71% 0 55 22 6823796,80 0 0 1 0 0 0

t-1 40 -1,04% 0 56 21 7992100,00 0 0 1 0 0 0

t0 40 0,06% 1 57 23 9484655,00 0 0 1 0 0 0

t+1 40 -0,15% 1 58 23 10750482,00 0 0 1 0 0 0

t+2 40 -2,47% 1 59 24 10382163,00 0 0 1 0 0 0

t+3 40 -1,57% 1 60 24 9814901,00 0 0 1 0 0 0

t+4 40 1 61 0 0 1 0 0 0

t-2 41 25,96% 0 35 46 7649000,00 0 0 1 0 0 0

t-1 41 19,48% 0 36 46 7483000,00 0 0 1 0 0 0

t0 41 21,27% 1 37 42 6254000,00 0 0 1 0 0 0

t+1 41 25,89% 1 38 41 6168000,00 0 0 1 0 0 0

t+2 41

1 39

0 0 1 0 0 0

t+3 41

0 0 1 0 0 0

62

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+4 41

0 0 1 0 0 0

t-2 42 21,41% 0 4 62 2289470,00 0 0 0 0 0 1

t-1 42 -8,50% 0 5 65 2130538,00 0 0 0 0 0 1

t0 42

1 6 63 2102288,00 0 0 0 0 0 1

t+1 42 -3,01% 1 7 63 2698691,00 0 0 0 0 0 1

t+2 42

1 8

0 0 0 0 0 1

t+3 42

1 9

0 0 0 0 0 1

t+4 42 0 0 0 0 0 1

t-2 43 -1,72% 0 23 221 10320114,00 0 0 0 0 1 0

t-1 43 0,38% 0 24 203 16109763,00 0 0 0 0 1 0

t0 43 -3,25% 1 25 198 23651849,00 0 0 0 0 1 0

t+1 43 -13,72% 1 26 187 12804529,00 0 0 0 0 1 0

t+2 43 -1,87% 1 27 166 16918340,00 0 0 0 0 1 0

t+3 43

1 28

0 0 0 0 1 0

t+4 43

1 29

0 0 0 0 1 0

t-2 44 4,20% 0 21 425 29538262,00 77877086,00 0 0 0 1 0 0

t-1 44 8,43% 0 22 481 34725690,00 100265692,00 0 0 0 1 0 0

t0 44 -2,49% 1 23 496 33901682,00 80715931,00 0 0 0 1 0 0

t+1 44 2,97% 1 24 510 37501684,00 93774770,00 0 0 0 1 0 0

t+2 44

1 25

0 0 0 1 0 0

t+3 44

1 26

0 0 0 1 0 0

t+4 44 0 0 0 1 0 0

t-2 45 29,79% 0 41 87 15739000,00 1 0 0 0 0 0

t-1 45 19,95% 0 42 92 19668000,00 1 0 0 0 0 0

t0 45 19,57% 1 43 93 21566000,00 98606000,00 1 0 0 0 0 0

t+1 45 20,53% 1 44 99 27026000,00 109407000,00 1 0 0 0 0 0

t+2 45 25,33% 1 45 113 30684000,00 124884000,00 1 0 0 0 0 0

t+3 45

1 46

1 0 0 0 0 0

t+4 45

1 47

1 0 0 0 0 0

63

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t-2 46 11,82% 0 44 119 22373226,27 86913366,39 0 0 0 1 0 0

t-1 46 15,88% 0 45 125 19066666,00 73116587,00 0 0 0 1 0 0

t0 46 9,65% 1 46 133 18942113,00 74311143,00 0 0 0 1 0 0

t+1 46 7,77% 1 47 83 21488313,00 73049833,00 0 0 0 1 0 0

t+2 46 7,32% 1 48 139 24066236,00 97354448,00 0 0 0 1 0 0

t+3 46 0,69% 1 49 142 20248865,00 80575010,00 0 0 0 1 0 0

t+4 46 1 50 0 0 0 1 0 0

t-2 47 10,86% 0 48 63 9887542,00 0 0 0 1 0 0

t-1 47 9,83% 0 49 63 9790178,00 0 0 0 1 0 0

t0 47 9,69% 1 50 63 9598139,00 0 0 0 1 0 0

t+1 47 8,41% 1 51 63 10573631,00 0 0 0 1 0 0

t+2 47 3,98% 1 52 65 10810748,00 0 0 0 1 0 0

t+3 47

1 53

0 0 0 1 0 0

t+4 47

1 54

0 0 0 1 0 0

t-2 48 5,26% 0 2 184 30002000,00 80055000,00 0 1 0 0 0 0

t-1 48 3,14% 0 3 239 27540000,00 71612000,00 0 1 0 0 0 0

t0 48 -11,72% 1 4 245 33728000,00 57726000,00 0 1 0 0 0 0

t+1 48 6,89% 1 5 215 34580000,00 75043000,00 0 1 0 0 0 0

t+2 48

1 6

0 1 0 0 0 0

t+3 48

1 7

0 1 0 0 0 0

t+4 48 0 1 0 0 0 0

t-2 49 11,14% 0 16 1315 306321000,00 260042000,00 1 0 0 0 0 0

t-1 49 0,34% 0 17 1400 316212000,00 249351000,00 1 0 0 0 0 0

t0 49 -4,07% 1 18 1379 310913000,00 214788000,00 1 0 0 0 0 0

t+1 49 3,06% 1 19 1443 347272000,00 218934000,00 1 0 0 0 0 0

t+2 49

1 20

1 0 0 0 0 0

t+3 49

1 0 0 0 0 0

t+4 49

1 0 0 0 0 0

t-2 50 33,61% 0 31 83 3969837,00 10016876,00 0 0 0 0 0 1

64

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t-1 50 25,79% 0 32 75 4431842,00 9010320,00 0 0 0 0 0 1

t0 50 30,93% 1 33 71 3830224,00 9296436,00 0 0 0 0 0 1

t+1 50 30,91% 1 34 72 4302769,00 10547623,00 0 0 0 0 0 1

t+2 50 29,81% 1 35 71 4445013,00 10555586,00 0 0 0 0 0 1

t+3 50

1 36

0 0 0 0 0 1

t+4 50 1 37 0 0 0 0 0 1

t-2 51

0 20 92

0 0 0 0 0 1

t-1 51 12,97% 0 21 97 8871021,00 0 0 0 0 0 1

t0 51 9,40% 1 22 124 8215703,00 0 0 0 0 0 1

t+1 51 9,11% 1 23 121 8020852,00 0 0 0 0 0 1

t+2 51 10,89% 1 24 110 10992411,00 0 0 0 0 0 1

t+3 51 9,22% 1 25 110 10942847,00 0 0 0 0 0 1

t+4 51

1 26

0 0 0 0 0 1

t-2 52 4,31% 0 12 34 32057030,00 0 0 1 0 0 0

t-1 52 28,03% 0 13 36 45936024,00 45566828,00 0 0 1 0 0 0

t0 52

1 14 41 33173154,00 0 0 1 0 0 0

t+1 52 10,76% 1 15 40 35294179,00 39484977,00 0 0 1 0 0 0

t+2 52 9,35% 1 16 35 37443691,00 38903679,00 0 0 1 0 0 0

t+3 52

1 17

0 0 1 0 0 0

t+4 52 0 0 1 0 0 0

t-2 53 8,43% 0 36 141 11863199,00 26331671,00 0 0 0 1 0 0

t-1 53 4,20% 0 37 144 12432567,00 34327993,00 0 0 0 1 0 0

t0 53 -9,31% 1 38 149 12137314,00 29204459,00 0 0 0 1 0 0

t+1 53 0,68% 1 39 131 12568350,00 0 0 0 1 0 0

t+2 53

1 40

0 0 0 1 0 0

t+3 53

1 41

0 0 0 1 0 0

t+4 53

0 0 0 1 0 0

t-2 54 5,47% 0 11 618 62368932,00 190041633,00 0 0 0 1 0 0

t-1 54 7,76% 0 12 628 74572137,00 218243786,00 0 0 0 1 0 0

65

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t0 54 10,18% 1 13 672 79093026,00 294253965,00 0 0 0 1 0 0

t+1 54 10,56% 1 14 704 80922313,00 322982165,00 0 0 0 1 0 0

t+2 54

1 15

0 0 0 1 0 0

t+3 54

1 16

0 0 0 1 0 0

t+4 54 0 0 0 1 0 0

t-2 55 1,99% 0 88 506 115865000,00 87480000,00 0 0 0 0 1 0

t-1 55 2,25% 0 89 473 119795000,00 94374000,00 0 0 0 0 1 0

t0 55 2,18% 1 90 474 147014000,00 90410000,00 0 0 0 0 1 0

t+1 55 1,66% 1 91 503 164249000,00 111622000,00 0 0 0 0 1 0

t+2 55 2,09% 1 92 526 154352000,00 113093000,00 0 0 0 0 1 0

t+3 55

1 93

0 0 0 0 1 0

t+4 55

1 94

0 0 0 0 1 0

t-2 56 8,33% 0 3 98 22510000,00 47161000,00 0 0 0 1 0 0

t-1 56 4,42% 0 4 102 25823000,00 51214000,00 0 0 0 1 0 0

t0 56 1,48% 1 5 101 21071000,00 40405000,00 0 0 0 1 0 0

t+1 56 -5,82% 1 6 87 19431000,00 26806000,00 0 0 0 1 0 0

t+2 56 2,75% 1 7 19385000,00 32383000,00 0 0 0 1 0 0

t+3 56

1 8

0 0 0 1 0 0

t+4 56 0 0 0 1 0 0

t-2 57 3,78% 0 10 164 500550913,70 162756606,20 0 0 0 0 0 1

t-1 57 2,90% 0 11 189 566127038,90 175534957,20 0 0 0 0 0 1

t0 57 2,10% 1 12 174 514364153,82 178651256,42 0 0 0 0 0 1

t+1 57 3,88% 1 13 170 527695000,00 180800000,00 0 0 0 0 0 1

t+2 57 5,13% 1 14 172 524268000,00 188592000,00 0 0 0 0 0 1

t+3 57 4,11% 1 15 155 568825000,00 186997000,00 0 0 0 0 0 1

t+4 57 4,15% 1 16 162 551615000,00 187591000,00 0 0 0 0 0 1

t-2 58 16,38% 0 43 89 3854767,64 0 0 0 0 0 1

t-1 58 10,37% 0 44 89 3704013,73 13302048,83 0 0 0 0 0 1

t0 58 10,38% 1 45 79 3704119,21 13301763,86 0 0 0 0 0 1

66

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+1 58 2,85% 1 46 82 4475039,06 17322882,96 0 0 0 0 0 1

t+2 58 5,00% 1 47 86 5293642,31 0 0 0 0 0 1

t+3 58 6,60% 1 48 85 5776987,91 0 0 0 0 0 1

t+4 58 11,53% 1 49 85 6354761,74 0 0 0 0 0 1

t-2 59 10,16% 0 93 246 12497731,21 0 0 0 0 0 1

t-1 59 8,78% 0 94 258 12360790,03 0 0 0 0 0 1

t0 59 10,15% 1 95 263 14191741,75 0 0 0 0 0 1

t+1 59 12,96% 1 96 261 17070996,00 0 0 0 0 0 1

t+2 59 10,04% 1 97 271 21304189,00 0 0 0 0 0 1

t+3 59 9,13% 1 98 297 19834492,00 23969366,00 0 0 0 0 0 1

t+4 59 5,34% 1 99 301 21201120,00 24390903,00 0 0 0 0 0 1

t-2 60 27,76% 0 51 33 5371196,66 0 0 0 0 1 0

t-1 60 26,47% 0 52 44 6756913,93 0 0 0 0 1 0

t0 60 32,67% 1 53 40 6865863,33 0 0 0 0 1 0

t+1 60 38,70% 1 54 36 7328842,00 0 0 0 0 1 0

t+2 60 31,70% 1 55 41 8149014,00 0 0 0 0 1 0

t+3 60 52,23% 1 56 49 16791280,00 0 0 0 0 1 0

t+4 60 6,30% 1 57 33225379,00 0 0 0 0 1 0

t-2 61 10,31% 0 7 29921109,60 0 0 1 0 0 0

t-1 61 9,45% 0 8 148 35585875,47 0 0 1 0 0 0

t0 61 7,24% 1 9 167 35172392,50 0 0 1 0 0 0

t+1 61 10,15% 1 10 176 38161672,06 0 0 1 0 0 0

t+2 61 10,23% 1 11 183 42192301,00 41166596,00 0 0 1 0 0 0

t+3 61 3,52% 1 12 187 63966857,00 0 0 1 0 0 0

t+4 61 3,23% 1 13 205 66064040,00 0 0 1 0 0 0

t-2 62 21,53% 0 17 86 5572281,54 0 0 1 0 0 0

t-1 62 16,51% 0 18 104 6389686,08 0 0 1 0 0 0

t0 62 12,40% 1 19 98 7176543,06 0 0 1 0 0 0

t+1 62 5,62% 1 20 95 6342046,00 0 0 1 0 0 0

67

Year Firm_id ROA Certified Age Employees Balance sheet Turnover C19-21 C26-28 E F H M+N

t+2 62 13,28% 1 21 88 6152197,00 0 0 1 0 0 0

t+3 62 -3,38% 1 22 85 6628248,00 0 0 1 0 0 0

t+4 62 -0,30% 1 23 80 6057758,00 0 0 1 0 0 0

t-2 63 8,47% 0 99 250 22090211,26 50218735,14 0 1 0 0 0 0

t-1 63 -2,93% 0 100 197 30240712,80 44654738,81 0 1 0 0 0 0

t0 63 3,64% 1 101 193 27428154,94 52446202,97 0 1 0 0 0 0

t+1 63 -2,92% 1 102 36014000,00 59897000,00 0 1 0 0 0 0

t+2 63 3,93% 1 103 214 35888000,00 78033000,00 0 1 0 0 0 0

t+3 63 4,41% 1 104 222 44738000,00 73344000,00 0 1 0 0 0 0

t+4 63 0,41% 1 105 219 41112000,00 68864000,00 0 1 0 0 0 0

t-2 64 9,24% 0 12 262 9081006,53 26914018,79 0 1 0 0 0 0

t-1 64 5,54% 0 13 265 10553038,91 28136102,40 0 1 0 0 0 0

t0 64 8,32% 1 14 262 12217023,54 28905055,43 0 1 0 0 0 0

t+1 64 6,95% 1 15 269 11140994,50 31549985,06 0 1 0 0 0 0

t+2 64 7,42% 1 16 100 11538994,04 33153983,62 0 1 0 0 0 0

t+3 64 -4,35% 1 17 255 9487477,00 25284338,00 0 1 0 0 0 0

t+4 64 6,85% 1 18 213 10525986,00 27145993,00 0 1 0 0 0 0

t-2 65 17,97% 0 23 57 10290678,06 0 1 0 0 0 0

t-1 65 15,27% 0 24 102 16878991,92 0 1 0 0 0 0

t0 65 21,61% 1 25 102 20013990,08 0 1 0 0 0 0

t+1 65 5,88% 1 26 47 19106444,00 0 1 0 0 0 0

t+2 65 8,79% 1 27 51 20815531,00 0 1 0 0 0 0

t+3 65 21,52% 1 28 58 29902756,00 0 1 0 0 0 0

t+4 65 24,96% 1 29 67 20828880,00 0 1 0 0 0 0

68

Appendix C – Hausman test fixed effects model versus random effects model

Hausman fixed random

---- Coefficients ----

(b) (B) (b-B) sqrt(diag(V_b-V_B))

fixed random difference S.E.

Certified (-.0142246) (-.0303877) (0.0161631) (0.0097525)

LNAge (-.112242) (0.019697) (-.131939) (0.0763516)

LNEmployees (-.0133716) (-.0093531) (-.0040185) (0.015797)

LNBalances~t (-.1065825) (-.0315693) (-.0750132) (0.0459043)

LNTurnover (0.0982308) (0.0186826) (0.0795482) (0.0325442)

Year3dum (-.0094718) (-.0062266) (-.0032452) (0.009752)

b = consistent under Ho and Ha; obtained from xtreg

B = inconsistent under Ha, efficient under Ho; obtained from xtreg

Test: Ho: difference in coefficients not systematic

chi2(6) = (b-B)'[(V_b-V_B)^(-1)](b-B)

= 9.95

Prob>chi2 = 0.1269

(V_b-V_B is not positive definite)

69

Appendix D - Breusch and Pagan LM test to test for random effects

Breusch and Pagan Lagrangian multiplier test for random effects

ROA[Firm_id,t] = Xb + u[Firm_id] + e[Firm_id,t]

Estimated results:

Var sd = sqrt(Var)

ROA .0107438 .1036524

e .0068855 .0829792

u .002585 .0508427

Test: Var(u) = 0

chibar2(01) = 14.63

Prob > chibar2 = 0.0001

70

Appendix E – Cronbachs Alpha test

Test scale = mean(unstandardized items)

Item-test Item-rest Average inter-item

Item Obs Sign correlation correlation covariance Alpha

ROA 329 - 0,0535 0,0421 0,0383342 0,5768

Certified 429 + -0,2037 -0,2352 0,0389467 0,5886

LNAge 430 + 0,0317 -0,0316 0,0360336 0,6058

LNEmployees 313 + 0,5611 0,4756 0,0193107 0,4514

LNBalances~t 331 + 0,6314 0,5391 0,0188863 0,4622

LNTurnover 200 + 0,8035 0,6453 0,0222592 0,4672

Economicsi~n 455 + -0,1099 -0,1297 0,0397422 0,5847

D1 455 + 0,0044 -0,0146 0,0382745 0,5726

D2 455 - -0,0029 -0,0267 0,0381717 0,5761

D3 455 - 0,1573 0,1293 0,0364977 0,5661

D4 455 + 0,1534 0,1194 0,0337947 0,5499

D5 455 + 0,0238 0,0049 0,0397354 0,5837

D6 455 - 0,0465 0,0172 0,0389215 0,5848

Test scale 0,0335928 0,5776

71

Appendix F – Fixed effects estimations

Model 1 Model 2 Model 3

VARIABLES ROA ROA ROA

Certified

-0.014 -0.015

(0.017) (0.018)

LNAge -0.151* -0.112 -0.125

(0.063) (0.077) (0.080)

LNEmployees -0.012 -0.013 -0.012

(0.019) (0.019) (0.019)

LNBalancesheet -0.108* -0.107* -0.110*

(0.049) (0.049) (0.049)

LNTurnover 0.097* 0.098* 0.099*

(0.038) (0.038) (0.038)

Economicsituation -0.007 -0.009 -0.007

(0.025) (0.025) (0.026)

Certified_Z1Employees

-0.005

(0.012)

Certified_Z1Balancesheet

0.010

(0.011)

certified_Z1turnover

-0.011

(0.053)

Constant 0.763 0.593 0.680

(0.731) (0.759) (0.772)

Observations 183 183 183

R-squared 0.111 0.116 0.121

Number of Firm_id 43 43 43

Standard errors in parentheses

*** p<0.001, ** p<0.01, * p<0.05, t p<0.10