People or a tool? Does ICT lead to higher labor productivity?

People or a tool? Does ICT lead to higher labor

productivity?

"Analyzing the impact of ICT innovations on labor productivity of Dutch manufacturing firms and the moderating effect of organizational innovation on this relation."

Master's thesis by Tim Janssen Master Thesis Business Administration

Specialization ‘Innovation and Entrepreneurship' | Nijmegen Radboud University

Name: Tim Janssen

Student number: s4595696

Supervisor: dr. P.M.M. Vaessen 2nd _{examiner: dr. ir. L.J. Lekkerkerk}

i |

P a g e

Abstract

This research examines the influence of ICT on labor productivity and checks how organizational innovations influence this relationship. The purpose of this research is to find to what extent and how ICT influences labor productivity and to what extent this relationship is influenced by organizational innovations. This in order to ultimately contribute to the discussion in the literature about the relationships between these variables. There are various tendencies, who have each their own point of view on the relationship between the variables ICT, organizational innovation, and labor productivity. This study examines these relationships via a mixed-method study to find if there are relationships and how these variables are related. Quantitative analyses of the EMS (2015) and semi-structured interviews with five manufacturing companies provide some interesting findings. First of all, ICT innovations do not influence labor productivity. Moreover, concerning the moderating variable organizational innovation, only one part of this innovation form influences the relationship between ICT and labor productivity. Structural organization innovation, in addition to ICT innovation, does strengthen the influence on labor productivity. According to the quantitative analysis, procedural organizational innovation does this not. Besides, the quantitative analysis revealed that structural and procedural innovation together do not strengthen the influence of ICT on labor productivity. However, the qualitative analysis reveals that procedural does strengthen this relationship as well. No qualitative support is found for the significant interaction effect of ICT and structural innovation. Moreover, interviewees argue also that there is no relationship between ICT innovation and labor productivity. Therefore, the results reveal that investing in ICT innovations does not lead to higher productivity, but at the moment, a company decided to invest in these innovations, then additional structural organizational innovations are desirable.

Key words: Labor productivity; ICT innovation; Organizational innovation; Structural organizational innovation; Procedural organizational innovation.

ii |

P a g e

Preface

This document contains the completion of my Master's thesis and the completion of my Master's Innovation and Entrepreneurship. Moreover, hopefully, this Master’s thesis is the completion of my period as a student. The thesis is about how the investment in ICT innovations is related to a development in labor productivity, and to what extent this form of innovation should be supplemented with organizational innovations. Because I use ICT tools daily during my work, this thesis offered me theoretical and practical insights. By finishing this thesis, I finish an eventful study year. Writing this research was a bumpy route to an end product. These bumps were caused by the challenges of writing a Master Thesis and the developments in the world due to the Corona Crisis. Contact was made more difficult by these developments in the world, which meant that matters such as classroom feedback meetings and interviews had to be solved creatively. This makes me extra proud that I completed the thesis before the deadline of the 15th _{of June. Without the support of several people, this was not possible, so I would like to}

thank several people for their contribution and support to this Master Thesis.

First, I would like to thank my supervisor dr. Peter Vaessen. With his support, guidance, and critical feedback, he helped me to take my Thesis to the highest possible level. He helped me, including during the Corona period, to critical review and where possible, improve my writings. I would like to thank my second examiner, dr. ir. L.J. Lekkerkerk, as well for his feedback on my proposal. This was also very helpful. Moreover, I would especially thank my thesis circle group. Cas, Daphne, and Bas kept helping me during my whole thesis period. They helped me stay motivated and supported me when I had questions. I am also very grateful to the companies who participated in my research. Without them, it was not possible to conduct a qualitative analysis.

Lastly, I would like to thank my family and friends who supported me in carrying out this research. I have not always been the nicest person during this period. This must also have been hard for my girlfriend, so I want to thank her for the support and relaxation while writing my master thesis. Thanks a lot to everyone!

Tim Janssen

Table of content Abstract ... i Preface ... ii 1. Introduction ... 1 2. Theoretical framework ... 3 2.1Labor productivity ... 3 2.2 Innovation ... 4

2.2.1 Product and process innovation ... 5

2.2.2 Technological innovation ... 5

2.2.3 ICT innovation ... 5

2.2.4 Organizational innovation ... 6

2.2.5 Organizational innovation forms ... 7

2.3 Relationship between ICT innovation and labor productivity ... 8

2.4 Moderating relationship of organizational innovation ... 10

2.5 Conceptual model ... 13

3. Methodology ... 14

3.1 Research design ... 14

3.2 Quantitative research method – EMS (2015) ... 15

3.2.1 Operationalization ... 16

3.3 Qualitative research method ... 18

3.3.1 Operationalization ... 19

3.4 Validity and reliability ... 19

3.5 Ethics ... 20

4. Quantitative study and results ... 21

4.1 Sample characteristics ... 21

4.2 Construction of the variables ... 22

4.2.1 Construction of dependent variable ... 22

4.2.2 Construction of independent variable ... 23

4.2.3 Construction of moderating variables ... 23

4.2.4 Construction of control variables ... 24

4.3 Univariate analysis ... 25

4.3.1 Dependent variable labor productivity ... 25

4.3.2 Independent variable ICT innovation ... 25

4.3.3 Moderating variables ... 25

4.3.4 Control variables ... 26

4.4 Bivariate analysis... 27

4.5 Multivariate analysis ... 30

4.5.1 Testing of assumptions ... 30

4.5.3 Moderating relation of structural or procedural organizational innovation ... 31

4.5.4 Moderating relation of structural and procedural organizational innovation ... 32

4.5.5 Other findings... 32

4.6 Additional regression ... 34

4.6.1 Autonomous relationship of the organizational innovation forms ... 34

4.6.2 Interaction Firm size with ICT and Organizational innovation ... 35

4.7 Summary of the results ... 38

5. Qualitative study and results ... 39

5.1 Labor productivity ... 39

5.2 Relationship ICT innovation – Labor productivity ... 40

5.2.1 New production techniques/machines ... 41

5.2.2 Affected people ... 42

5.2.3 Efficiency ... 43

5.2.4 The effect of ICT innovation on labor productivity ... 43

5.3 Influence of organizational innovativeness ... 44

5.4 Summary of the main findings ... 46

5.5 Additionally qualitative insights to the quantitative insights ... 47

6. Conclusion and discussion ... 49

6.1 Conclusion ... 49

6.1.1 Summary of the research ... 49

6.1.2 Answer to the main question ... 50

6.2 Discussion ... 51

6.2.1 Hurdles towards (positive) influence of ICT on labor productivity ... 51

6.2.2 Contradictory findings about procedural organizational innovation ... 52

6.2.3 Reinforcing effect of structural and procedural innovation together ... 53

6.2.4 Additional findings ... 53 6.3 Theoretical implications... 54 6.4 Practical implications ... 54 6.5 Limitations ... 55 6.6 Future research ... 55 Reference list ... 57 Appendices ... 60 Appendix 1: EMS-Survey (2015) ... 60

Appendix 2: Quantitative operationalization... 66

Appendix 3: Tree-structures ... 69

Appendix 4: Informed consent ... 71

Appendix 5: Interview guide ... 72

Appendix 6: Code Scheme ... 77

Appendix 7: Construction of Organizational Structural Innovation ... 78

Appendix 9: Checking of Assumptions linear regression ... 81

9.1 Assumption of the first regression ... 81

9.2 Assumptions of the second regression ... 83

9.3 Assumptions of the third regression ... 85

Appendix 10: Linear Regression ... 87

10.1 Linear regression 1 ... 87

10.2 Linear regression 2 ... 88

10.3 Linear Regression 3 ... 89

Appendix 11: Interview transcripts ... 90

Interview 1 ... 90

Interview 2 ... 98

Interview 3 ... 105

Interview 4 ... 116

Interview 5 ... 123

Appendix 12: Interview analysis ... 131

General Questions: ... 131 Interview 1: ... 133 Interview 2 ... 138 Interview 3: ... 143 Interview 4: ... 151 Interview 5 ... 156

1 |

P a g e

1. Introduction

Most of the people have heard about the Democratic pre-election in Iowa 2020. It was the first time that the Democratic party used an app to count the votes, but this did not go entirely smoothly. Instead of speeding up the voting count process, it finally took almost a week before the final result was announced because some users did not understand the app, and the app also did not function as they expected. Even afterward, the result was challenged by various politicians. The goal of the Democratic party was to speed up productivity with the use of ICT, instead of this, productivity declined. A possible explanation for this was that the app had not been tested beforehand.

This decrease in productivity is a very extreme phenomenon, but the growth in productivity has been declining for years (Roelandt, 2019). The data from The Conference Board (2019) showed that labor productivity grew by around 4% per year in the 1970s, and in 2010 this percentage lay below 1%. Despite some revivals, the pace of growth is slowing (Andrews, Criscuolo & Gal, 2016). One of these revivals was in the mid-1990s. Labor productivity growth rate (LPGR) increased in this period because of technological innovation and semiconductor manufacturing (Anderson & Kliesen, 2006). The Dutch ICT expenses (goods and services) grew from 44,9 billion in 2010 to 60.6 billion in 2018 (CBS, 2018). In line with the situation in the 90s, you would, therefore, expect that LPGR increases. However, this is not the case, and therefore, there are many opinions about the influence of ICT innovation on labor productivity. This research is interested in these different opinions and the impact on the productivity of Dutch manufacturing firms.

One of the movements in the literature is skeptical about the influence of ICT innovation on LPGR. Solow (1987, p. 36) mentioned in mid-80's "you can see the computer age everywhere but in the productivity statistics”.

Researchers like Cowen (2011), Gordon (2012), and Jones (2012) stated that we have fallen back to more normal growth of ICT and that it will become more expensive to build on the innovations made in the mid-90s and the beginning of this century. These authors provided arguments like, that people have fully utilized the easily accessible ICT innovations (Cowen, 2011) and that more and more knowledge is required to use ICT innovations (Jones, 2012). So, this movement sketches a predominantly negative picture of the current influence of ICT innovation on the LPGR.

On the other hand, there is a movement that states that ICT innovation has a more positive influence on the LPGR in the (near) future. Brynjolfsson and McAfee (2011) stated that we live in a society where we use more and more information and communication technology, and our innovation capabilities increased (Lopez-Nicolas & Meroño-Cerdán, 2011). As a result, the ICT revolution will continue rapidly, and ultimately, this will create new business models and a new wave of productivity growth (Brynjolfsson and McAfee, 2011; Mokyr, 2014). So, these authors are more positive about the influence of ICT innovation on LPGR and even argued that it would lead to a new wave of productivity growth.

Mokyr (2014), who is mainly positive about the influence of ICT innovations, added to his positive view that several barriers could hinder the positive impact of ICT innovation on the LPGR. He argued that it could lead to a shift in focus on ICT innovation to other side issues, or as he calls it, barriers. One of these barriers is an inadequate institutional distribution of research fundings, which leads to incremental instead of radical innovations. Other barriers are the rise of cybercrime and internal resistance entrenched by different interests. These barriers could lead to extensive regulations (Mokyr, 2014). Parham (2002) stated that the policy of companies should focus on competition and flexibility instead of a strong focus on ICT-regulation. This showed

2 |

P a g e

that ICT innovation not only influences labor productivity, but the organizational circumstances also play a role in this (De Ridder, 2019; Inklaar, 2019; Roelandt et al., 2019).

However, in the existing literature, no agreement has been found on the extent to which the influence of ICT innovations on labor productivity depends on organizational innovation. Roelandt et al. (2019) believed that when ICT does not coincide with organizational innovation, ICT cannot have a strong positive effect on labor productivity. Inklaar (2019) stated in a contrast that intangible complementary investments for most companies do not give the expected return. De Ridder (2019), is even more resolute and argued that supplementary expenditures, in addition to ICT, actually curb productivity.

The goal of this research is to contribute to the discussion about the influence of ICT innovations on labor productivityand what role organizational innovation plays in this since there is currently no consensus about these relationships in the literature. The contribution is twofold. Firstly, to see whether and how ICT innovations influence labor productivity. Second, to discover to what extent this relationship is influenced by organizational innovations. By contributing to one of the tendencies in literature, hopefully, someday, a clear vision on this subject will arise. In order to contribute to this discussion, the following question will form the basis for this research

"To what extent and how does ICT innovation influence labor productivity of Dutch manufacturing firms, and to what extent and how is this relationship influenced by organizational innovations?"

As mentioned above, there is no agreement about the influence of ICT on labor productivity. Solow's (1987) statement will be taken as the starting point for this research. In the studies that were carried out, no consensus was found about the influence of ICT. Therefore, this study presents an overview of these different opinions, in order to ultimately take a position in the discussion about the influence of ICT on labor productivity. This assumption, based on the existing literature, will be checked with a mixed-method study. This study is, therefore, relevant because it gives a current picture of the relationship between ICT and labor productivity and therefore contributes to the discussion in the literature about this relationship. Moreover, it looks not only to the quantitative aspect but also to the qualitative aspect of the relationship.

Based on this contribution of this research, managers/policymakers have more information to make responsible choices about whether they want to innovate in ICT, and it can also be determined in advance whether they should invest in additional organizational innovations. These insights could prevent a situation such as that with the Democratic pre-elections.

The data for this research will be retrieved from the European Manufacturing Survey 2015 (EMS, 2015). Based on the results of this survey and six additional semi-structured interviews, an attempt will be made to answer the research question. The research will be structured as follows:

In the second chapter, the theoretical constructs will be elaborated. This chapter gives an overview of the concepts in this research and how the different relationships are measured based on existing literature. The third chapter is about the methodology in this research. It will be elaborated on how the mixed-method approach is used to check whether the hypotheses are significant and how the qualitative insights are obtained. In chapters 4 and 5, the quantitative and qualitative results are presented. Based on these results, an answer to the main question will be drawn in chapter six. This chapter contains the answers to the main- and sub-questions and an assessment of the quality of the research. Also, the limitations of this research and a recommendation for future research will be made in this chapter.

3 |

P a g e

2. Theoretical framework

This chapter is about the theoretical framework of this research. The goal of this chapter is to get an understanding of the constructs ICT innovation, organizational process innovativeness, and labor productivity, based on existing literature. Firstly, these concepts will be described in this chapter. In the literature, there is no agreement about the influence of ICT innovation on labor productivity. Therefore both views will be discussed. By choosing one of these opposing views, a hypothesis will be formed about the influence of ICT innovation on labor productivity. Besides, scientists also disagree about whether or not to invest in organizational processes, in addition to ICT innovations, to increase labor productivity. These opposite views will be discussed in this chapter, and afterward, hypotheses will also be formed. Finally, based on these hypotheses, a conceptual model will be derived. The conceptual model gives a visual overview of the expected relationships in this research.

2.1 Labor productivity

Labor productivity forms the dependent variable of this research and will be influenced by the independent variable ICT innovation. Organizational process innovativeness will have a moderating effect on this relation.

Productivity is a strong concept to measure the level of competitiveness of a firm, and it indicates the economic conditions and national power of a country (Mačiulytė-Šniukienė & Gaile-Sarkane, 2014). Krugman (1992) and Van Ark & Monnikhof (2000) have also concorded that an increase in labor productivity and efficiency are causes for the rise of the living standard. So this indicates the importance of labor productivity for companies, but also for the society as well. The Fraunhofer Institute has defined labor productivity as the added value per employee. So, the revenue of a company minus the total purchasing divided by the number of employees (Kinkel, Rieder, Horvat & Jäger, 2015).

The Dutch Centraal Bureau voor de Statistiek (CBS, 2015, p.4) has argued in their reports about the productivity in the Netherlands between 2002 and 2014 that you can subdivide the reasons underlying labor productivity into three parts. The first part is capital deepening. The second part is labor composition. The last factor that forms the productivity is multifactor productivity. Oliner and Sichel (2000) already have considered these factors to determine the development of the manufacturing output.

According to Oliner and Sichel (2000, p.11), capital deepening is the "amount of capital per hour worked." The capital to labor ratio can increase by a rise in capital stock or due to a lower number of employees. It is possible to assign any type of capital investment to a particular factor, for example, ICT investments, to see what the influence of this specific investment on productivity (Oliner and Sichel, 2000).

The labor composition factor is more abstract than the capital deepening. Oliner and Sichel (2000) have given a general definition of labor composition, namely "the quality of the delivered labor." Therefore, the definition of labor composition provided by Jablonski, Rosenblum, and Kunze (1988), gives a clearer picture on which the quality of labor depends. They have argued that the labor composition, and therefore the quality of work, depends on the work experience, educational level, and gender of an employee. Hence, the change in the labor composition of a company can influence the productivity of a company.

The last factor that determines labor productivity is multifactor productivity, according to CBS (2015). Although, other authors (e.g., OECD, 2005) have considered this as a stand-alone form of labor productivity rather than an explanatory factor for labor productivity. This factor is the remaining part of the labor productivity

4 |

P a g e

development that is not due to the above factors. Oliner and Sichel (2000) have called it a "catch-all" for organizational and technological innovation, which let the output grow for a given number of input. Dutch CBS (2005) has added some factors to this, which determine multifactor productivity. They have stated that technological development, economies of scale, increasing utilization rates, and non-economic factors (for example, adverse weather conditions who have a negative influence on the agriculture) as well, impact multifactor productivity. This factor measures several things that influence labor productivity and develop itself in line with the development of labor productivity in general, with a growth rate below 1% in the last decade (Conference Board, 2019; OECD, 2019). It does not take into account one thing as a capital investment or the quality of the work, but it takes several things into account, which together influence labor productivity.

This study is interested in the influence of ICT innovation on labor productivity in general. Therefore, the definition of the added value per employee of the Fraunhofer Institute (Kinkel, Rieder, Horvat & Jäger, 2015)

serves as the basis for this research. Figure 1 gives an overview of the development of the labor productivity growth rate. Here is the declining growth development, as mentioned in the introduction, also visible. Compared to the growth in the 1970s, the growth is nowadays significantly lower.

Figure 1. Labor productivity development G7 countries (retrieved on 20-03-2020 from https://data.oecd.org

/lprdty/multifactor-productivity.htm OECD-data) 2.2 Innovation

This section provides the first picture of the concept of ICT innovation. The paragraph is, as it were, arranged in a funnel shape. First of all, it starts with a definition of the broad concept of innovation. Innovation can be divided into different forms, namely product, process, organizational and technological innovation (Armbruster et al., 2007). These four forms of innovation will be addressed after the broad definition of innovation. In the following paragraphs, the terms will be linked together, and further explanation will be given of how these terms ultimately form the independent variable ICT (process) innovation and organizational structural and procedural innovation. Investments in new technology can be seen as an innovation. Innovation in itself is a broad concept with several meanings. One of these general definitions is 'a new idea, method, or device: novelty' (Merriam-Webster's Collegiate Dictionary). A more specific description of innovation from the OECD (2005, p.46) is "the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organizational method in business practices, workplace organization or external relations." It encompasses the whole process from idea to launch (Cooper, 2001). So, innovation can take various forms, such as product and process innovation or organizational and technological innovation, and it develops over time (Armbruster et al., 2007). Armbruster et al. (2007) have created a quadrant (figure 2) in which all these different forms of innovation are discussed.

4,1% 4,8% 1,7% 2,1% 3,4% 2,9% 0,8% 2,6% 1,0% 2,7% 2,7% 1,9% 2,4% 1,8% 2,8% 1,5% 0,2% 2,0% 2,9% 2,0% 2,1% 1,1% 0,4% 2,4% 0,6% 0,0% 0,9% 1,0% 0,3% 1,3% 0,1% 0,0% 1,0% 2,0% 3,0% 4,0% 5,0% 6,0% 1 9 7 1 1 9 7 2 1 9 7 3 1 9 7 4 1 9 7 5 1 9 7 6 1 9 7 7 1 9 7 8 1 9 7 9 1 9 8 0 1 9 8 1 1 9 8 2 1 9 8 3 1 9 8 4 1 9 8 5 1 9 8 6 1 9 8 7 1 9 8 8 1 9 8 9 1 9 9 0 1 9 9 1 1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1 2 0 1 2 2 0 1 3 2 0 1 4 2 0 1 5 2 0 1 6 2 0 1 7 2 0 1 8

5 |

P a g e

According to Maranville (1992), a product is innovative when it helps to fulfill new market needs or existing market needs in a different way. Besides a product, a process can also be innovative (Edquist, Hommen & McKelvey, 2001). A process is innovative when it implements a new production or delivery method, or that it improves an existing production or delivery method (OECD, 2005, p.17). Evangelista and Vezzani (2010) have added to this that process innovation leads to a competitive advantage via efficiency and productivity gains acquired by the introduction of better performing ways of producing (current) products. So, this is one of the distinctions that are made in the literature about the different forms of innovation. Product and process innovation are closely related to technological (product/process) innovation (OECD, 2005, p. 47). Therefore, in the following sub-paragraph, technological innovation will be discussed.

2.2.2 Technological innovation

Technological innovation activities are all the steps, including investments in new knowledge, which actually, or are intended to, lead to the implementation of technologically new or improved products and processes (OECD, 2002, p.18).Technological innovation can, therefore, be product-oriented and process-oriented (OECD/Eurostat, 1997).

According to the OECD (2002, p. 9), "A technological product innovation is the implementation/ commercialization of a product with improved performance characteristics such as to deliver objectively new or improved services to the consumer". Examples of technological product innovation are the use of ABS braking systems in car manufacturing or, more generally, the development of microprocessors. A technological product innovation is aimed at the consumer.

Technological process innovation is, according to the OECD (2002, p.32), "the adoption of technologically new or significantly improved production methods, including methods of product delivery." This kind of innovation encompasses the change in the organization of production, change in equipment, or a combination of these two, and can be obtained from new knowledge. The goal of this new method(s) is to increase the delivery and/or production efficiency of existing products, or the target can be to create new ways of producing and deliver new products (OECD, 2002). Technological process innovations are, for example, innovations that are aimed to lower the production costs, to lower the disposal costs, or to increase the ability to use cheaper materials. A tangible example is the use of an ERP system for the automatic handling of logistical, administrative, and financial business processes. The next paragraph starts with the elaboration of the independent variable ICT innovation.

2.2.3 ICT innovation

As mentioned above, this study focuses on a particular part of technological innovation, namely ICT-innovation, and even more specific, in an organization. ICT is a well-known concept and is widely implemented in the manufacturing world. Therefore, this study focuses on technological innovation because almost every company has to deal with ICT. ICT has developed itself actively over the years, starting with the introduction of processors in the 1970s. The development of the first computers followed reasonably quickly. Then, in the early 1990s, the creation of the internet was added. The beginning of this century was marked by the introduction of mobile phones

6 |

P a g e

and other portable devices (van Liempt, 2006), and nowadays, we mostly work in the cloud instead of on physical equipment. These developments also had their influence on labor productivity (figure 1), with peaks in the mid-70s and early 90s. ICT is nowadays used to collect, generate, communicate, administer, and distribute information. It consists of software, hardware, and networks, and also media for storage, processing, transmission, collection, and presentation of data (Sarkar, 2012, p. 32).

In the previous section, a distinction was made between technological product and process innovation. In some cases, it is difficult to make this hard distinction, as well as for ICT innovations. Some authors have argued that the difference between an ICT process or product innovation is arbitrary and can change over time (Edquist, Hommen, & Mckelvey, 2001). This is an assumption that is not supported by this thesis. An ICT innovation can be a product innovation for a particular company initially and a process innovation for another company. An example of this is the introduction of Office 365. For Microsoft, this innovation is an ICT product innovation, which it wants to bring to the market, to sell to their customers. Companies buy this product from Microsoft and implement this in their organization. This implementation can lead to the acceleration of the primary production process, for example, because construction teams can share ideas faster with each other through Microsoft Teams. So for the companies who buy Office 365 from Microsoft, this is an ICT process innovation which they use (mostly) inside their organization.

The context, therefore, determines to what extent an innovation is an ICT product or process innovation. Because it was decided to only look at ICT innovations within organizations and because this thesis uses the EMS (2015) as a data source, where it is difficult to find a context, it is decided to see all technological process innovations from the EMS, which match with the definition of ICT of Sarkar (2012), as ICT innovations in this study. That is why in figure 2 technical product innovation is marked with a dotted box, because in this research it really only includes new ICT products for use within the organization. Think of a self-developed app within an organization to keep track of stocks.

2.2.4 Organizational innovation

According to Armbruster et al. (2007), you have, besides technological innovation, also non-technical innovation. Organizational innovation is the form of non-technical innovation that focuses on the processes in an organization (Armbruster et al., 2007). Organizational innovation influences the social system of an organization. This system includes the rules, structures, and procedures that are linked with organizational members and are interrelated with the environment and these organizational members (Cummings & Srivastva, 1977). Organizational innovations do not lead to new products or services but indirectly influence new services or products or the way of producing products (Kimberly & Evanisko, 1981). Evangelista and Vezzani (2010, p. 1259) have specified this by stating that organizational innovation has to do with objectives like "the reduction of the time needed to respond to customer or supplier needs" and of "improvement of the quality of goods or services." Examples of organizational innovations derived from the work of Matsunaga (2019, p.6) are "the first introduction of management systems, business reengineering, lean production, and quality management systems." So this type of innovation mainly deals with people and the organization of the work (OECD, 2005).

7 |

P a g e

Note: Reprinted from: Patterns of organisational change in European industry (porch): Ways to strengthen the empirical basis of research and policy, by Ambruster, H., & Fraunhofer-Institut für System- und Innovationsforschung, 2007, Innovation papers, 46, p. 20

2.2.5 Organizational innovation forms

Several authors have used (e.g., Vaessen, Ligthart & Dankbaar, 2012; Armbruster et al., 2007) the definition developed by Damanpour and Evan (1984, p. 393) as a definition for organizational innovation; "The managerial and organizational systems and practices that are new for and put to use in the organization."

The definition of organizational innovation in this study is more specific and based on the definition that Armbruster et al. (2007) have used in their work. They have seen organizational innovation as a new organizational process or organizational structure that is developed and implemented in the organization, intending to offer employees and customers more efficiency and flexibility. However, this is still a broad and general definition of what organizational innovation means. Therefore, Armbruster et al. (2007, p. 20) have specified it more by dividing it into structural organizational innovation and procedural organizational innovation. Both types of innovation influence another factor in the organization. Structural organizational innovations influence and improve responsibilities, command lines, and information flows. Moreover, it also influences the distribution of power, the separation between functions, and the divisional structure in organizations. Examples of this type of innovation are cross-functional teams or the division of labor (see figure 3). Procedural organizational innovations lead to a change or an implementation of new procedures or processes and affect the operations, routines, or processes of a company. It can influence the speed and quality of production. Examples of this are just-in-time concepts or quality circles or KANBAN-principles (see figure 3). Armbruster et al. (2007) have made further a distinction between intra-organizational innovation and inter-organizational innovation, in other words, organizational innovation inside the organization (intra) or between (inter) companies. This study focusses on the procedural and structural organizational innovations that occur inside (intra) the company.

Figure 3. Organizational innovation forms

Note Reprinted from: Patterns of organisational change in European industry (porch): Ways to strengthen the empirical basis of research and policy, by Ambruster, H., & Fraunhofer-Institut für System- und Innovationsforschung, 2007, Innovation papers, 46, p. 21

8 |

P a g e

In the literature, there are divergent opinions about the influence of ICT innovation on labor productivity. By studying the literature, a distinction can be made between two movements. One movement which argues that ICT-innovation has a positive influence and the one who states that the impact of ICT-ICT-innovation has flattened or even that ICT-innovation has no influence on labor productivity. These two movements will be explained in this paragraph, and ultimately a point of view will be chosen that forms the basis for the first hypothesis of this study. "You can see the computer age everywhere but in the productivity statistics" (Solow, 1987, p.36). This statement is made by Solow in 1987 about the relationship between technological ICT development and labor productivity. He has looked to the US manufacturing industry in the 1980s and formed a personal statement about the developments in this industry. This view about the impact of ICT on labor productivity will be the dominant relationship in this study. In the years after that Solow made this statement, several authors have agreed with him, but there was also evidence that ICT has a positive influence on labor productivity. This positive movement was strongly represented at the time of the rise of the internet and the beginning of this century. Parham (2002) is one of the authors who has disagreed with Solow's statement. He has studied the ICT influence 20 years after Solow and thus formulates his statement based on newer developments. Parham (2002) has studied the economic figures from the 1990s in Australia and compared them to the US, and he has mentioned that the rise of the internet, computers, and telecommunication systems has led to a significant change in education, businesses, healthcare, and other aspects of society. He has stated that ICT-innovations create platforms that enable people to communicate and process data quicker. This leads to lower costs for saving and spreading of information, and, also to faster communication, information processing, and coordination during the production process. This faster communication and lower costs to use data lead to lower production costs and, therefore, an increase in labor productivity (Parham, 2002).

Brynjolfsson and Hitt (2000) have conducted a similar study, but then on a firm-level. They used seven case examples of big companies like General Motors and Dell to generate qualitative evidence, and they have studied several quantitative studies to support these case examples. They have discovered, in line with Parham (2002), that ICT innovation leads to lower costs of the above-stated factors. They have added to this that ICT investments lead not only to lower costs but also, more importantly, to the improvement of the quality of products. Parham (2002) and Brynjolfsson and Hitt (2000) have written their studies at the beginning of the century, at the time of the introduction of portable devices. Nowadays, the opinion of most authors has changed. Brynjolfsson and McAfee wrote a new book in 2011 in which they recalled the influence of ICT at the beginning of this century. On this point, however, they have concluded that the positive impact of ICT at the turn of the century declined and that a positive wave is likely to return in the future. Another example of a positive author who has become more skeptical about the influence of ICT on labor productivity is Mokyr (2014). He was predominantly positive about the influence of ICT innovation on labor productivity at the beginning of this century, and he studied the development of ICT in the US over the years based on economic numbers. He has argued, however, that the positive influences of ICT developments on labor productivity are increasingly complicated by barriers. These barriers, like for example, cybercrime, lead to extensive regulations. This shifts the focus of companies toward ICT-regulation instead of innovation. This means that the positive influence of ICT on labor productivity is inhibited by these barriers.

9 |

P a g e

Therefore, the view of the positive authors changed during current times. There are nowadays several authors who are less positive. Professor Cowen (2011) from the University of George Mason has called the influence of ICT in his pamphlet about the development of the US economy the "great stagnation" of the US economy. For decades the US people have caught all the low hanging fruit of technological innovation to increase labor productivity, among other things. People expected this fruit to occur again and again, but this fruit has slowly disappeared in the last 40 years. ICT has become more and more complicated; the purchase of it has become more expensive, and the implementation has become more complicated. The net positive effect has, therefore, disappeared.

Next to the argument that we have caught all the more easily accessible ICT-innovations, ICT also requires knowledge, since technology depends on the understanding that the user has about this. Jones (2012) has investigated a data set of 2.1 million utility patents in the US and clustered this based on the grant date, application year, and the technological category to find the fundamental aspects of the technological progress. He has discovered that knowledge becomes more and more scattered among different people within the society/company. Everyone has experience in his or her part, and the specialization of employees increases. Due to technological innovation, more knowledge is also required. This means that there are more employees, with all their specialization of a part of the product, are needed to produce a (technical) product or to execute a process. Peter Drucker (1991) has written earlier that technology in itself does not increase labor productivity; it is about the people and the way they use the technology. So a situation does arise in which labor becomes more knowledge-intensive, and this knowledge is spread over more people. Production, therefore, becomes more dependent on the expertise of employees, and this knowledge is spread over a larger group of employees. It has thus become more challenging to increase labor productivity.

Nkama (2014) have argued, moreover, in his research that there is a negative relationship between ICT investments and the development of labor productivity. This research was conducted among 81 enterprises in Cameroon, of which 41 companies were active in the manufacturing sector. He has also indicated that ICT innovation is very dependent on the available knowledge in an organization. Due to the low spread of knowledge about ICT in Cameroon, it is even the case that ICT innovation has a significant negative effect on labor productivity. Arendt & Grabowski (2018) did comparable quantitative research, among 820 SME's in Poland, where knowledge about ICT is more spread among the population. They have looked to the influence of ICT usage on labor productivity and revealed that there was no significant effect of ICT usage on labor productivity. This was because the ICT innovations must be accompanied by organizational changes and specialized personnel, which is already scarce.

Taking this together, ICT innovations go hand in hand with extra work such as cybersecurity, and it ensures that more and more knowledge, which is spread over different people, is needed to produce. Like Drucker (1991) mentioned, "whether tools help productivity or harm, it depends on what people do with them." As a result, in this study, it is assumed that the development of labor productivity is not significant linear related to ICT innovations. Therefore, the following hypothesis has been formed:

H1: "In manufacturing firms, the development of labor productivity is not significantly linearly related to the adopted ICT-innovations."

10 |

P a g e

Organizational innovation is the moderating variable in this study. The influence of organizational innovations on the relationship between ICT innovations and labor productivity will be examined. As well as the contrasting opinions about the impact of ICT innovation on labor productivity, there is also no clear consensus in the literature as to whether organizations should (simultaneously) invest in organizational processes if they invest in ICT or not. One group of authors states that a simultaneous investment leads to an (additional) increase in labor productivity. In contrast, other authors declare that a simultaneous investment does not bring the desired or even no benefits. This paragraph will start with the movement in the literature who are negative about the influence of organizational innovations

"Focusing on one thing at a time" is a well know Dutch saying. This saying fits with the movement in the literature which states that organizational and technological innovation do not coincide, or at least not at the same time. This group of authors (e.g., Inklaar, 2019; De Ridder, 2019) believes that investing in technological innovations does not go hand in glove with investments in organizational innovations. All companies can invest in a computer with more or less the same functions. This technological innovation is for every company the same. Only a small part of the companies can also invest in an adequate trainings-program for their employees. Every trainings-program differs from other programs, and therefore, also the quality. A computer is a fixed set with parts and software with possibly some add-ons, but training can be given in infinite ways and, therefore, cannot be copied (Inklaar, 2019). Andrews, Criscuolo, and Gal (2016) have studied the balance sheets of companies with a 2-digit code between 5 and 82 (+20 employees) in 24 OECD countries to find more evidence at firm-level about the factors behind the global productivity slowdown. In line with Inklaar (2019), they have concluded that only a small set of companies, the productivity leaders, are able the invest (enough) in their complementary organizational innovations (e.g., training) to profit from the technological innovation as well (Andrews, Criscuolo, & Gal, 2016).

Besides, De Ridder (2019) has used two datasets of 1,087,726 France firms about their financial performance in the period between 1994 and 2016 and found that growing ICT-investments lead to a shift in the cost structure of an organization. The higher investments in complementary intangible inputs, and the primary investment in ICT-innovations, lead to a change from variable cost towards more fixed costs. These higher fixed costs lead to a higher share of these fixed costs in the total costs of a company. These costs influence the costs per hour worked, and therefore, have a negative influence on labor productivity.

However, the sociotechnical system theory has argued that any modification of the technological systems of an organization requires a change in the organizational system, this because a new technical system leads to new demands (Cooper & Foster, 1971). So, according to this theory, organizational innovation is a precondition for technological innovation. Lam (2005, p. 115) has reflected this in her own words by stating that "the ability of an organization to innovate is a precondition for successful utilization of . . . and new technologies."

Damanpour and Evan (1984) have conducted a quantitative study based on 85 surveys at public libraries, and they have found that restructuring and organizational innovation lead to a regulatory climate with coordination and cooperation. This leads to a climate in the organization where technological innovations can be adopted and utilized. Due to the organizational innovations, people communicate and corporate better which each other (Damanpour and Evan, 1984), and it is necessary to provide as many people as possible with the knowledge about ICT innovations. Employees' understanding of ICT innovation is required to benefit from ICT innovation, ultimately (Drucker, 1991). Moreover, Teece (2010) have studied about how to organize your business model to

11 |

P a g e

implement innovations properly. He has concluded that if a company wants to profit from technological innovation, then a company should adopt new organizational methods, models, and forms because, otherwise, technological innovation will certainly not be successful. These new organizational forms, models and, methods are, therefore, at least as important or even so more important as technological innovation.So it is widely accepted in the literature that the overarching concept of organizational innovation promotes the influence of technological innovations.

One of the first academic foundations for this positive influence of organizational innovations on the relation between technological ICT investments and the development of labor productivity was laid by Damanpour and Evan (1984). Still, more empirical studies followed, as mentioned above. Some studies specify the influence of the two organizational innovations forms, as indicated by Armbruster et al. (2007). However, research containing this specific split is still scarce. Gunday, Ulusoy, Kilic, and Alpkan, (2011) have agreed with Damanpour and Evan (1984), and they have found, based on an empirical study covering 184 manufacturing firms

in Turkey, that only one round of a restructuring of your organization and other organizational changes are not

enough to realize technological innovation. According to them, technological innovation must be provided with a structural organizational innovation, so that all employees change their way of working together. Structural change of the way people work together in an organization is necessary to implement technological innovation successfully. The case example of General Electric (GE), revealed by Hamel (2006), is another strong example of the strong influence of structural organizational innovation. GE decided to restructure the teams within the technological innovation department and to add management layers to this. Besides, they introduced an ICT-tool to spread information. As a result, knowledge about various technological innovations was spread more quickly and accurately within the organization, which ultimately led to GE being able to develop these technological innovations further. This increases the number of successful patent applications for new technological innovations, higher than ever before. So, according to these authors, structural organizational innovation could positively stimulate the effect of an ICT innovation on labor productivity or, in this case, successful production. Therefore, the following hypothesis is formed:

Hypothesis 2a: "The greater the number of (simultaneously) adopted structural organizational innovations, the stronger the positive relationship between ICT-investments and labor productivity."

Besides the influence of structural organizational process innovation, there is also empirical evidence for the positive impact of procedural organizational innovation on the successful implementation of technological innovations. Cozzarin (2016) has done a longitudinal study among 2990 Canadian firms in 2009 and 2617 in 2012 to find the influence of organizational innovations on technological innovations and found a positive relationship. He has checked which organizational innovations were used and found that 58.6% used new business practices for organizing procedures, in other words, procedural organizational innovation. These procedural organizational innovations, like quality circles, ensure that employees feel more responsible and involved. As a result, they accept technological innovations more quickly, so that employees start working faster with an innovation. Combating resistance in the organization, for example, through quality circles, ensures that employees start using a technological innovation faster and better (Blaga & Jozsef, 2014). From this follows the following hypothesis:

12 |

P a g e

Camison (2014) has measured, in his quantitative study among 114 Spanish industrial firms, organizational innovativeness as a combination of innovations in workplace organization and business practices, so structural and procedural organizational innovation together. He has concluded that this form of innovation favours the development of technological product and process innovation. These organizational innovations support the dissemination of knowledge about a technological innovation in an organization and decrease resistance to innovation because employees are involved in innovations. So, in other words, Camison (2014) has argued that structural and procedural innovation together favours the development of ICT innovation because this is a form of technological innovation. Based on this, and the sociotechnical system theory in general, the following hypothesis is formed:

Hypothesis 2c: "The greater the number of (simultaneously) adopted procedural and structural organizational innovations, the stronger the positive relationship between ICT-investments and labor productivity."

13 |

P a g e

Based on the literature which is used for this study, the conceptual model (figure 4) given below is developed. This conceptual model should guide in answering the main question: To what extent and how does ICT innovation influence labor productivity of Dutch manufacturing firms, and to what extent and how is this relationship influenced by organizational innovations? The relationships tested in this study are visualized in the model and will be tested in order to answer the main question ultimately.

The autonomous relationship exists between ICT-innovation and labor productivity. In this study, this relationship is seen as a so-called null relationship. According to, e.g., Arendt & Grabowski (2018), there is no significant relationship between ICT innovation and the development of labor productivity. Explanations for that are that ICT innovations go hand in hand with extra work such as cybersecurity (Mokyr, 2014). It ensures that more and more knowledge, which is spread over different people, is needed in the production process (Jones, 2012). Therefore, in line with the hypothesis, it is stated in the conceptual model that ICT-innovations do not influence labor productivity.

The moderating factor in this research is organizational innovation. This study assumes that organizational innovation has a positive effect on the relationship between ICT-innovations and labor productivity. Thereby, the split is made between structural and procedural organizational innovation. It is assumed that structural (Gunday, Ulusoy, Kilic, and Alpkan, 2011), as well as procedural organizational innovation (Cozzarin, 2016), have a positive autonomous influence on the relationship between ICT innovation and labor productivity. Moreover, structural and procedural organizational innovation together also influences the relationship between ICT innovation and labor productivity positively (Camison, 2014). Therefore, at the moment that there is organizational innovation, in addition to ICT-innovation in an organization, the effects of ICT-innovation on labor productivity will be positively influenced

.

.

Figure 4. Conceptual model

ICT Innovation

Labor productivity

Structural

organizational

innovation

Procedural

organizational

innovation

Organizational innovation

14 |

P a g e

3. Methodology

This chapter is about the methodology that was used in this study. This chapter forms the basis of the methodological justification. Several decisions and further explanations will be elaborated in this chapter. In the first paragraph, the research design will be presented. As mentioned, this will be mixed-method research and, therefore, in paragraph 3.2, the quantitative approach will be elaborated, and in paragraph 3.3, the qualitative method approach. In these two paragraphs, the operationalization of the variables will be elaborated as well. The fourth paragraph will be about the validity and reliability of this research. The chapter ends with a section about research ethics.

3.1 Research design

This research focused on a specific quantitative analysis of the EMS (2015). The goal of this research was to find to what extent and how the different variables in this research were related to each other. Therefore this specific quantitative part is used to answer the question to what extent the variables are related. Besides, these quantitative results of the analysis are discussed with several managers and employees, via interviews, to obtain qualitative insights from people in practice. This gave answers about how these variables are related. So, therefore, a mixed-method research design was used for this study. In the perception of the researcher, the combination of these research methods was the most appropriate research design to receive answers on the questions of this research and to achieve the goal of this research. A mixed-method study can help to get more substantial evidence and additional insights, and it increases the generalizability (Creswell & Zhang, 2009).

The quantitative part of this research ensured that it was possible to look if and to what extent there is a relationship between ICT and labor productivity and the influence of organizational innovation on this relationship. The qualitative part of this research could respond to this by further, more in-depth, explaining the relations, and helping with the interpretation of the connections. This is all in line with Yin (2014), who has stated that mixed-methods enables researchers to gather a richer and more substantial array of evidence. So first of all, quantitative research is conducted to check if and to what extent there are relationships in this research. After this quantitative part, qualitative research is undertaken to serve as an additional information source and to check and complement the quantitative analysis.

The data for the quantitative part was gathered from the European Manufacturing Survey (EMS), which was carried out in the period between 2012-2014 – by Dr. Peter Vaessen and Dr. Paul Lightart (in the Netherlands). The goal of this survey was to get insights into the efforts of Dutch industrial companies to modernize their production and business processes. This survey has been distributed to 18 countries in Europe, however only the Dutch data is used for this study.

Besides the quantitative data, the qualitative data was gathered through six semi-structured interviews with three managers who implement technological innovation in a manufacturing company and three production employees who have to work with this innovation daily. This number with this split into functions was a target number, but due to the corona crisis, the quantity and separation could deviate from the predetermined plan. This split was made to deepen further the opinion of the persons who introduced the innovation as well as the opinion of the person who encountered the innovation. The answers of the managers and production workers on the interviews were used to deepen out the relations, which followed from the analysis of the EMS-data. Also, missing results or unclear results could be further elaborated with these interviews. Ultimately, these two methods are used

15 |

P a g e

to get a full picture of the reality to form a well-founded answer to the research questions. In the following paragraph, the quantitative research method will be described more in detail and the qualitative research method in paragraph 3.3.

3.2 Quantitative research method – EMS (2015)

The data for the quantitative research was conducted, as mentioned earlier, from the EMS (2015). This survey was conducted around 18 countries in Europe and was coordinated by the German Fraunhofer Institute for Systems and Innovation Research. Central in this survey was the technological and organizational innovations in manufacturing companies, which aim to improve the performance of these companies. For this research, only the data from Dutch manufacturing companies, with more than ten employees, was used. This Dutch data was collected by Peter Vaessen and Paul Lighthart, who are working for the Radboud University Nijmegen. The analysis started with the univariate analysis to visualize and mainly summarize the data. The second step was the bivariate analysis, which checked the correlation between the independent, dependent, and moderating variables. Besides, the bivariate analysis checked also whether there is multicollinearity. This means two explanatory variables are strongly correlated, i.e., at least one of them can be predicted based on the model. If this is not the case, then a multivariate analysis will be deployed to test the hypothesis (Field, 2013). This was done based on a regression analysis. Regression analysis is an appropriate method to examine the relationships between dependent and independent variables (Yin, 2014). The software program SPSS was used to conduct this quantitative research.

The data that was used for this analysis is a so-called 'secondary data', which was gathered by the EMS. The advantages of this secondary data are the time-saving and cost-reduction of the collection process because the data no longer has to be collected since it is already available. Also, bigger and more qualitative databases can be gathered, which would not be feasible for an individual researcher. A negative aspect of secondary data is that data can be outdated, or the information is collected for another research with another purpose (Secondary Data - Meaning, its advantages, and disadvantages, n.d.). Therefore, it is essential to compare the different research goals of the EMS and this study. The EMS looked at technological and organizational innovations that aim to improve the performance of the company. The purpose of this study was to clarify the relation between ICT-innovation and labor productivity. The objectives of the various studies, in general, were therefore not so different that the use of secondary data could be seen as problematic.

For the autonomous relationship in this study, linear regression was used. Linear regression is a form of regression analysis, which tests the relationship between a (multiple) independent and a dependent variable (Yin, 2014). This method was in line with the goal of this study since this research was interested in the relationship between the independent variable ICT-innovation with the dependent variable labor productivity. Besides this relationship, there was also the moderating variable organizational innovation. This variable was split into structural and procedural organizational innovation. The influence of these two moderating variables was tested by forming three interaction variables. These interaction variables were variables that were created in SPSS by multiplying the independent variable with the moderating variables. The analysis started with testing the autonomous relation, then the interaction variables were added to the model, and then there was checked whether interaction variables were significant. If that is the case, then you can speak about a significant moderating effect (Hair et al., 2013).

16 |

P a g e

In order to make the quantitative analysis possible, a link was made between the theoretical concepts described in chapter 2 and the questions from the EMS. Items from the EMS were linked to the dependent, independent, moderating, and control variables. The full overview of this operationalization is presented in appendix 2. The following paragraph elaborates on how these different variables are formed.

3.2.1.1 Labor productivity

The dependent variable of this research was labor productivity. There was checked whether ICT-innovation influences the development of labor productivity. In this thesis was chosen to measure labor productivity as the added value per employee. Therefore, this required three indicators, namely the revenue per year, the purchases of a company per year, and the number of employees. To come to the variable labor productivity, the calculation was as follows:

(Revenue per year (21A-I) –total purchases (21E)) / Number of employees (21B) Question number Indicator

21A-I Revenue per year of the company

21B Number of employees

21E Total purchases

Table 1. Labor productivity

3.2.1.2 ICT innovation

ICT innovation formed the independent variable of this research. As mentioned in chapter two, all ICT related process and product innovations within an organization were seen as ICT innovations. In the EMS, there were several questions about ICT. For the operationalization of ICT innovation, 22 items from question 8 were used, whereby it was checked whether an ICT innovation was implemented and was upgraded. Additionally, three sub-questions of sub-questions 3,5 and 7 were used. Table 2 gave an overview of the indicators that were used to operationalize this concept.

Question number Indicator

3H-I Graphical view of work processes

5.2C-I E-learning

7A-I Specified IT-safety rules

8.1C-I/IV Systems who stop machines by underproducing

8.1D- I/IV Automated systems for energy-efficient production

8.1N- I/IV Machine2Machine communication

8.1O- I/IV Cloud-computing

8.1P- I/IV Digital production planning

8.1Q- I/IV Real life production management planning

8.1R- I/IV Digital supply chain management

8.1S- I/IV RFID/warehouse management

8.1T- I/IV Tablets for installations or regulations of machines

8.1U- I/IV PLM systems

8.1W- I/IV Tablets for digitalizing work-schemes and instructions Table 2. ICT innovation

17 |

P a g e

The moderating variable organizational innovation was split into two forms of organizational innovation; structural innovation and procedural innovation, based on the split that was made by Armbruster et al. (2007). Table 3 gives an overview of the operationalization of structural organizational innovation. Table 4 provides an overview of the operationalization of procedural organizational innovation.

Structural organizational innovation

This form of organizational innovation focused on influencing and improving responsibilities, command lines and information flows, and also the distribution of power, separation between functions, and the divisional structure in organizations. So, this innovation form focused, in particular, on how people worked together. Indicators from question 3 about organizational concepts were used to operationalize structural organizational innovation

.

Question number Indicator

3A-I Regulations of equipment and storage

3B-I Standardizing work instructions

3E-I The functional layout of production units

3Q-I Standardizing methods for functionals designs Table 3. Structural organizational innovation

Procedural organizational innovation

Procedural innovation focused more on how people worked, what were the procedures that belonged to their job. This form of process innovation had more influence on the way people worked and led to a change or an implementation of new procedures or processes and affect the operations, routines, or methods of a company. Table 4 gives an overview of which indicators are used for the operationalization of procedural organizational innovation.

Question number Indicator

3C-I Job enrichment production employee

3D-I Improvement of internal logistics

3F-I Demand-driven production

3G-I Prescriptive methods to reduce lead time

3J-I Methods for operation management

3K-I Methods for continues improvement

3L-I Certified energy management

3M-I Product-life cycle analyzing methods

5.1A The standard amount of training days

5.1B Number of training days per year

5.2D-I On-the-job training (job rotation) Table 4. Procedural organizational innovation

3.2.1.4 Control variables

Firm size, industry, firm age, level of education, and other technological innovations were used as control variables in this study. The control variables are used to check whether the generalizability of the relationships, which are