Stimulating Innovation in the Netherlands through Governmental Interference

(1)

Stimulating Innovation in the Netherlands

through Governmental Interference

- The concept of behavioural additionality -

Master’s Thesis

Written by

J. M. Tammeling

1

st

_{supervisor: Dr. F. Noseleit}

2

nd

_{supervisor: Prof. Dr. W. A. Dolfsma}

University of Groningen

Faculty of Economics and Business

Msc Strategy & Innovation

August, 2012

Gedempte Zuiderdiep 55a 9711 HB Groningen

06-30410072

(2)

This master’s thesis focuses on the influence of governmental support programmes on the innovative behaviour of small and medium enterprises (SME’s). The study is performed as a multiple case study in which six cases, consisting out of businesses participating in the Innovatie Krediet, are interviewed on the perceived influence of such a program on their innovative behaviour. Several dimensions of behavioural additionality are thereby addressed on three different organizational levels. Also, the perceived change in organizational routines is taken into account. The conclusions that can be drawn from this study are that a public support program like the Innovatie Krediet does not directly influence the innovative behaviour of companies, although it does have an influence on several other, more general aspects of business activities and behaviour.

(3)

- 3 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling

I » Introduction

This master’s thesis will focus on the additionality effects that may result from governmental intervention on the level of innovation of SME’s within the Netherlands. The concept of behavioural additionality, as main point of interest, will be explored in an empirical setting by conducting a multiple case study among six participants of the Dutch Innovatie Krediet program. First of all, the research topic will be elaborated upon, followed by a brief explanation of the Innovatie Krediet program. Finally the research problem and research questions are presented followed by the structure of the research.

I.1 Research topic

The main theme guiding this thesis is governmental policy on stimulating innovation and its accompanying effects. Before addressing the concept of behavioural additionality a brief explanation of the reasoning behind public support mechanisms is presented. The institution of public support policy related to innovation can take different shapes and may be driven by various reasons. Financial stimulants can be considered to be the most common. For example, by tax reductions on the use of R&D personnel or tax reductions on royalties and other innovation related profits (Owens, 2010). Another way of financial stimulation could be the issuing of subsidies or credit arrangements by the government. This kind of policy is mainly grounded on the theory of ‘market failure’ within innovation investment (Arrow, 1962). The reluctance of businesses to innovate creates a gap between the demand and supply of innovative activity. The inability or unwillingness within businesses to innovate may be the result of a lack of appropriability (Pisano & Teece, 2007), the perceived level of risk involved, or a lack of available resources. According to Görg & Strobl (2007) innovation activities therefore inherently entail some sort of market failure. Private returns will most likely be lower than social returns, which results in businesses innovating below a certain social optimal level. This provides an incentive for governments to design public support policy with regard to the stimulation of innovation.

(6)

- 6 -

the Netherlands. While most evaluation studies concerning this topic focus on direct measurable effects of public support (mostly input and output additionality), this thesis strives to investigate the less obvious effects, or effects which may not seem that important at first sight. Following the concept of ‘behavioural additionality’ (e.g. Buisseret et al. 1995; Georghiou, 2002; Clarysse et al. 2009) the alterations in firm behaviour due to public support mechanisms are subject of investigation. This concept indicates that firms participating in a public support program may be motivated to change their attitude and effort towards innovative activities as a result of the government intervention. Examples of behavioural additionality for instance may be the institution of a research strategy or the collaboration with new research partners. This thesis aims at assessing these alterations in firm behaviour by looking at two concepts in particular. First of all the different dimensions of behavioural change that may occur are used in order to detect behavioural additionality. These dimensions are related to specific business activities and are classified at different organisational levels. The second concept used to map behavioural additionality is the concept of organisational routines. Changing routines are considered to be quite helpful in order to detect behavioural changes in organisations. These two concepts will form the main theoretical components in order to assess the possible impact of behavioural additionality.

While the discussion of the concept of behavioural additionality had started around the late nineties, there has not been conducted much empirical research yet. According to Clarysse et al. (2009) there is a lack of empirical evidence supporting the theory of behavioural additionality. This research therefore aims at contributing to fill the gap between the theoretical concepts and the practical application of such a phenomenon. The focus of this thesis will be on public support with regard to stimulating the innovative capacity of SME’s. First of all because small and medium enterprises (SME’s) account for a large share of new product innovations (Audretsch, 2003) and secondly because of the fact that the larger part of the business environment in the Netherlands, almost 99% consists out of SME’s (MKB Servicedesk.nl, 2010). SME’s in Europe are considered to be businesses consisting of less than or equal to two-hundred fifty employees (Europa.eu, 2003).

I.2 Innovatie Krediet

(7)

- 7 -

projects which can be classified as highly risky. It is extremely difficult for these businesses to obtain external financing; therefore the government tries to support these initiatives by means of a loan. The executive organisation of the Innovatie Krediet is Agentschap NL (see Appendix Afor a more detailed explanation of the Innovatie Krediet).

I.3 Research problem and research questions

Typical evaluation studies on the effects of governmental support on innovation and R&D tend to focus solely on the quantitative and more obvious variables which are easy to translate into success or failure of a particular public funding program. The qualitative and perhaps less obvious variables are often left aside. This creates an interesting research field within economic evaluation studies with regard to R&D and innovation. It would be interesting to assess how visible these qualitative effects are in practice and how they affect organisational behaviour in innovating firms. This study tries to get hold of the effects of governmental financial stimulation on the organisational behaviour of small and medium enterprises in the Netherlands through conducting a case study among firms enrolled in the Innovatie Krediet program. To date, to the best of the author’s knowledge, this type of study has not been conducted for SME’s in the Dutch market yet. The goal of this study could therefore be described as: ‘to gain insight into the additionality effects of public funding in its attempt to stimulate the innovative behaviour of SME’s in the Netherlands, with special interest in the Innovatie Krediet program.’ The reason why these qualitative effects are an interesting subject of investigation is their assumed influence on organisational behaviour and their assumed enduring influence on the innovating capacity of participating firms. Assumed is, that when the organisational behaviour of a firm is altered due to a public funding program aimed at stimulating innovation, they are more likely to perform innovative activities in the future, therefore creating a self-sustaining cycle of innovation (Georghiou, 2002). However, the question is to what extent particular programs alter innovative behaviour within firms. The main research question or this thesis is therefore formulated as:

‘What is the influence of a program like the Innovatie Krediet on the innovative behaviour of Dutch SME’s?’

(8)

- 8 -

first of all it is assessed if any change has occurred at all. Then the aspects that have changed are discussed and finally the ways they have changed are evaluated.

The first question is:

 ‘To what extent does a program like the Innovatie Krediet lead to behavioural additionality?’

The aim of this question is to assess whether or not the concept of behavioural additionality appears to be present in the selected business cases. Is there, apart from input and output additionalities, a change in organizational behaviour noticeable with regard to innovative activities which can be related to the participation in a public support program?

The second question is formulated as follows:

 Which dimensions of behavioural additionality are most common in SME’s involved in the Innovatie Krediet program?

The concept of behavioural additionality can be divided into several types or dimensions presented in the literature. The goal of this question is to address which of these dimensions are affected and whether or not there is a dominant type or dimension of behavioural additionality present in the selected business cases. The dimensions are related to the level of persistency of the change on different organizational levels (project, R&D level, and organisational level).

The final question is stated as:

 In what way can the presence of public support alter the organisational routines of SME’s involved in the Innovatie Krediet program?

(9)

- 9 -

I.4 Structure of the research

(10)

- 10 -

II » Literature review

There is an abundance of literature available on the general stimulation of innovation and R&D. This theoretical reflection will present some of the theoretical concepts which can be derived from the literature. First of all innovation within SME’s will be discussed, hereby focusing on the possible problems that may arise and on the differences with larger firms. Secondly, there will be an elaboration on the justification of government interference in supporting innovative activities, mainly based upon the ideas of Nelson (1959) and Arrow (1962). This is followed by a brief overview of the differences between tax exemptions and direct funding. Fourthly, some of the possible drawbacks of public support and the crowding-out concept are discussed. The next part explains the possible additional effects, or so called additionalities, of public support programs. After this, the concept of behavioural additionality will be discussed into greater detail. The final two sections will discuss different dimensions of behavioural additionality and the link with organisational routines.

II.1 Innovation in SME’s

This thesis focuses on innovation within small and medium enterprises (SME’s). Innovation however can take place in any firm regardless of its size. From the literature several studies present differences in strategy, effect and the extent of innovation activity between small and large firms. Some of these findings are presented below.

(11)

- 11 -

large. They hypothesize that large firms should have the relative innovative advantage in concentrated markets imposing significant entry barriers. These firms are in a better position to exploit the gains from innovation in an industry that requires a high level of capital intensity. Furthermore, the ability of product differentiation through increasing advertising activities creates a greater innovative advantage for larger firms, while the smaller firms are limited in their innovative activity (Acs & Audretsch, 1987). Lastly, larger firms achieve an innovative advantage in markets with a high degree of concentration. According to Galbraith (1956) ‘innovation can only occur in the presence of such market power and only by a firm that has the resources which are associated with considerable size.’ The small firms have the relative innovative advantage in markets resembling a more competitive model. The opportunities for small firm innovation are the greatest in the early stages of the product life-cycle, in industries requiring a fairly high component of skilled labour. These conclusions were interpreted as evidence that the whole debate regarding which firm size is more beneficial to innovation is of no real use and the real issue is which industry characteristics favor either large or small businesses (Acs & Audretsch, 1987). In a later study (1991), Acs & Audretsch found that innovations increased less than proportionately with firm size in almost every one of the investigated sectors, with just a few exceptions. Cohen et al. (1987) and Cohen & Klepper (1996) in a later study came up with a similar conclusion. Not the actual firm size, but appropriability and technical opportunity at the industry level produced a great deal of the variance in R&D intensity among firms. Furthermore they posed that factors at the business unit level had much more impact on the probability of conducting R&D than at the firm level.

This limited review of the literature on the differences between innovating large and small firms indicates a shift over time from Schumpeter’s (1942) second hypothesis, in which he claims that small firms will not choose to participate in many innovative projects, towards the conclusion that small firms actually do engage in innovative activities. However they may encounter some barriers and constraints, depending on a specific industry or sector.

II.2 Justification of governmental interference

(12)

- 12 -

is difficult to keep the ‘produced’ knowledge secret to competitors or other possible recipients. When the knowledge cannot be appropriated fully by the investing company, the economic value and prospect of such an investment diminishes. This creates a reluctant attitude within firms towards the investment in R&D and other innovative activities. This is called the non-appopriability argument of this particular market failure (Arrow, 1962; Nelson, 1959). The second argument which is proposed by these authors is the level of uncertainty that accompanies the investment in R&D. Another characteristic of the investment in R&D are the divergences in riskiness. Investing in R&D brings along a level of uncertainty that cannot be compared to other types of investment (Hall, 2002). Especially at the beginning of a project where the optimal R&D strategy cannot be fully analyzed and the project has an options-like character. Therefore the final output of the investment cannot, or can only in a very limited way, be determined. This may also have the effect of limited external financing options, as external investors are reluctant to participate in a project of which the actual outcomes are yet to be determined. Furthermore, a distinct characteristic of the investment in R&D is that in most cases almost fifty percent of the investment is used up by wages/salaries of R&D personnel (Hall, 2002). They possess the knowledge necessary for development of the innovation, thereby acting as a knowledge base of the firm. This contributes to the level of uncertainty in the sense that the workforce has a certain level of mobility. This could mean not only that personnel could leave the company, but also that they could take with them the knowledge in which the company has invested so heavily. The last argument for the market failure in R&D investment is the indivisibility of knowledge. The generation of (basic) knowledge may require a minimum efficient scale which is larger than individual firms could possibly produce at. This could especially be the case for SME’s as they are more often limited in their range and availability of resources (Hall, 2002; Acs & Audretsch, 1987).

(13)

- 13 -

learning capabilities and to help them how to find complementary expertise in order to perform innovative activities (Falk, 2007).

II.3 Tax exemptions versus direct funding

(14)

- 14 -

the highest social return. This may be caused by the fact that government agencies make use of the ‘success-stories’ of technological advancement to improve their political image. Providing money to projects, which on first sight do not seem to deliver direct results, may be viewed as a waste of public money and may harm the image of the government (David et al. 2000). Therefore government agencies may be encouraged by direct funding to ‘pick the winners’, instead of relying on market selection (Guellec et al. 2003). This thesis focuses on the providing of direct funding due to the fact that subsidies or grants, contrary to tax reductions, are not designed for every business. Companies have to apply for a subsidy or grant, which takes time and effort, and their application has to be approved. Therefore it should, in theory, attract the businesses who are more genuinely involved in innovative activities, creating a more interesting research field.

II.4 Drawbacks of R&D subsidies

Theoretically justified optimal government interference does not by definition mean that the government can rectify the existing market failures in practice as well. Subsidies and subsidy programs also contain particular drawbacks and flaws. The most common ones will be discussed below.

First of all, Arrow (1962) makes reference to the moral dilemma issues. This involves changes in the incentives of the recipients of subsidies. He explains this concept by referring to the insurance branch. Arrow (1962) compares subsidies with insurance policies. Insurance policies cannot exceed a particular amount of insured value. This is because the insurance policy may eventually change the incentives of the insured. In the case of a fire insurance for example, a policy that exceeds a particular insured value may increase the incentive for carelessness or even arson (Arrow, 1962). The same goes for subsidies. When the amount of available public money for research becomes too high, the incentive for businesses to invest their private funds into research activities may become limited. Thus diminishing the actual effect for which subsidies are designed, that is to enlarge the involvement of private funds in particular investments.

(15)

- 15 -

funds, to invest elsewhere in the business. As a result the firm keeps a similar research budget, compared to the situation before government interference. This could be considered to be a crowding-out of private funds. One recent study within the context of R&D subsidies by Görg & Strobl (2007) also mentions this crowding-out effect. They state that there is the possibility that public funding will ‘crowd out’ private financing of R&D. This effect can occur because it is likely to be cheaper for firms to apply for a government subsidy than to raise funds in the capital market by themselves. This could mean that some projects may be publicly funded that would have been undertaken anyway, even without government support. While this may be due to informational deficiencies between the government and the firm, it could very well also be the result of policy-makers’ supporting those projects that are privately the most profitable in order to avoid being seen as wasting public funds (Görg & Strobl, 2007).

Another point of criticism on the distribution of public funding in the form of subsidies or grants could be placed with the risk assessment that the involved companies make. The availability of public money to firms could lead to a distorted vision of possible threats and risks. According to Georghiou (2004) firms could be encouraged to take risks that they cannot afford due to the external financial stimulant. Public intervention could therefore mean that firms are tempted to move beyond their competences or to undertake projects that were too risky (Georghiou, 2002).

The procedure followed in selecting the businesses that may be eligible for a subsidy can also create some problems. Subsidies are designed to deal with the market inefficiencies which prevent private R&D activities from producing up to the desired social optimum. Subsidies should therefore be granted to the best of the privately unprofitable innovation projects, in order to stimulate private R&D (Wallsten, 2000). However, as stated earlier, some policymakers tend to support those projects which are privately most profitable.

Government officials may select firms to participate in a particular program based on their likely success, in order to claim credit for the firms’ ultimate success. This is referred to as ‘picking winners’ (Jacobs & de Man, 1996). Thereby supporting the projects which would have been undertaken anyway, and thus stimulating the crowding out of private funds (Clausen, 2007).

II.5 The additionality concept

(16)

- 16 -

framework designed to explain the intended ‘catalyst effects’ of public support (Braein et al. 2002). From the literature three distinct types of additionality (input, output and behavioural) can be derived. Input additionality originated from several economic studies (Levy & Terleckyj, 1983; Klette & Moen, 1998) and describes the initial stage multiplier effects of R&D subsidies on the total amount of R&D expenditures (Clarysse et al. 2009). In other words, the extent to which public subsidies lead to more continued efforts in R&D processes by the company (Clarysse et al. 2009). This could mean for example an increase in man-hours dedicated to R&D or capital investments. This type of additionality can be described as non-persistent, and is most often measured on project level. After the project has finished this effect is no longer present. Within the input additionality concept, several possible scenarios may occur. In the first scenario the firm, for every Euro provided in public support, spends at least an additional Euro on the target activity. This would be the ideal scenario. Another scenario would be one in which the firm does not increase its expenditure on the target activity by an equal or greater amount than the received public support. In this case the firm diverts the public funds to enhance its profits (Georghiou, 2002). The last scenario would be one in which the firm accepts the public support for an activity it would have engaged in anyway, thereby spending the funds on another project. This could be considered as the ‘crowding-out’ of private funds.

(17)

- 17 -

Traditionally, the impact of R&D support is formulated in terms of input and output, the relationship between this input and output is considered to be quite straightforward (OECD, 2006). An increase in resources (by means of public support) at the input side can lead to increased private financial effort, which may lead to additional output that could not have been achieved without the support. This basic idea sounds logical at first sight. However, this simplified way of thought does not take into consideration the other possible effects that could occur from the support, besides additional input and output. These effects could be described by the concept of behavioural additionality. Behavioural additionality however, requires a deeper understanding of business innovation processes compared to input and output additionality, and internal processes of the firm should therefore be emphasized as well. The traditional view on input and output does not take into account these possible effects that may occur and treats the firm as a ‘black box’ (Hall & Maffioli, 2008). The effects on firm behaviour are not considered to be of interest within this view. The following figure aims to clarify the traditional view.

(18)

- 18 -

II.6 Behavioural additionality explained

From the literature several explanations of behavioural additionality can be derived. While they all have some common ground, there are clear and important differences to be found between them. Gök (2010) distinguishes between four general definitions of behavioural additionality. The first definition can be categorized as treating behavioural additionality as an extension of input additionality. This definition regards behavioural additionality as a rather simple concept, and only takes into account the fact that the presence of public support can alter a stop-go decision or alter the way in which the project was carried out. Projects can be carried out faster (acceleration additionality), on a larger scale (scale additionality) or on a wider range of applications (scope additionality) (Luukonen, 2000; Georghiou, 2002). In this view behavioural additionality is treated as a non-persistent effect and operating only at one point in time during the project. Furthermore, nothing is considered to spill-over or to endure beyond the duration of the support and the definition is confined to just the project, and in a limited way the R&D processes (Gök, 2010).

A second category of definitions presented by Gök (2010) is somewhat broader compared to the former. It defines behavioural additionality as the change in non-persistent behaviour related to R&D and innovation activities. This definition includes the change in behaviour of its agents, and looks beyond the extension of input and output additionalities. To put it differently: ‘behavioural additionality deals with the change in the way that organisations conduct their R&D due to governmental stimulation’ (Buisseret et al., 1995). This definition still limits this form of additionality to R&D processes and no link with persistent changes in behaviour that endure even after the project is finished is made (Gök, 2010).

The third category of definitions is rather similar to the second, except for the fact that in this category the persistent element of change in behaviour is taken into account. Behavioural additionality is defined as ‘the change in the persistent behaviour related to R&D and innovation activities (Gök, 2010). This definition emphasized on the fact that changes in behaviour may endure beyond the period of support or project.

(19)

- 19 -

change which has a more general character is recognizable in the increase of networking activities or so called ‘network additionalities’ (Falk, 2004).

Apart from the additional effects mentioned above, Kleer (2010) addresses another interesting fact. According to this author, the granting of subsidies by the government may also act as a signal for promising investments to private investors. Small banks or other private investors may find it difficult to assess the risks and potential benefits of a certain research project. The confidence of a private investor in a particular project could increase whenever active government support is involved. The government can resolve some of the information asymmetries that may exist between firms and private investors. In this perspective the granting of a subsidy by a government agency may act as an objective quality assessment for private investors. This effect could be labelled as ‘quality additionality’. While this concept may perhaps be more applicable to the behaviour of third parties, it could be considered to be an additional effect resulting from government intervention. The behaviour of the supported firm itself may not be directly affected, but it could affect the image of the firm. Firms might make use of their improved image to interact differently with third parties.

The discussion above indicates that behavioural additionality is a concept which is not universally defined. It varies from a narrow view in which input additionality seems to form the basis, to a broader view in which every behavioural change in the firm is attributed to government support. Also, the term behaviour may remain rather vague and is not easily measurable. Furthermore, the concepts of input, output and behavioural additionality may seem to overlap to some extent. However, the main difference between the behavioural and the input and output additionalities lies in the element of permanent (behavioural) change that occurs by means of engaging in a publicly funded research project. In most instances the effects of behavioural additionalities occur beyond the project level, thus making the effects more permanent. Therefore the ultimate goal of these behavioural additionalities should be to create a self-sustaining cycle of (successful) investment in innovation (Georghiou, 2002). In this way subsidized research helps to institutionalize certain routines and competencies within businesses with respect to further innovation and research. This may ease the process of innovation for these businesses in the future, striving to the point where no public funding is needed anymore.

II.7 Dimensions of behavioural additionality

(20)

- 20 -

 Project additionality (Davenport et al. 1998) – Indicates whether or not the project would have been continued without the public support. It concerns the alteration of the stop-go decision for the project.

 Acceleration additionality (Georghiou, 2002) – Whether the project has been executed faster, perhaps in order to meet a market window.

 Scope additionality (Georghiou, 2002) - Where the coverage of an project is expanded to a broader range of applications or markets than would have been possible without public support

 Scale additionality (Georghiou, 2002) - When the project is larger than it would otherwise have been as a result of public support

 Cognitive capacity additionality (Bach & Matt, 2002) – Assesses whether changes in the cognitive capacity of the management of recipients are realized due to public support.

 Network additionality (Falk, 2004) – Whether or not changes in networking activities have been realized due to public support.

 Quality additionality (Kleer, 2010) – Indicates if the public supports affects the image of the recipient organization, regarding for example investors.

Next to these types of behavioural additionality Larosse (2004) also describes several forms of additionality that could be considered complementary to the above.

(21)

- 21 -

level represents the level of persistency of the effect, by which the project level can be linked to less persistent effects and the system level to more persistent effects. The process level indicates the fact that the concepts of input, output and behavioural additionality are intertwined and should not be interpreted as totally isolated effects. The process level further refers to the phases in the innovation process (the ‘arrow of time’), a sequence of input, throughput (or internal processes), and output can be made (Larosse, 2004). It indicates the phase within the innovation process in which the effect takes place.

From the literature it can be concluded that several types of behavioural additionality are mentioned and that these can be categorized on different levels of aggregation. Combining the different views on the concept and taking into account the process and organizational level the following overview can be presented. This overview tries to include most relevant issues concerning the concept of behavioural additionality and will act as the general guideline for the continuation of this research. This research will assess these dimensions by addressing them at project, R&D and company level, thus emphasizing the organizational context in which the dimensions interact. This complies with the different definitions presented earlier from existing literature. Note that the first two definitions, the input and output additionality, are combined into the project level, as to their non-persistent character. A more detailed explanation of the different examined dimensions of behavioural additionality is presented in appendix B.

Overview of dimensions of behavioural additionality

Organizational level Phase

innovation process Project (non-persistent) R&D activities (persistent) Firm (persistent)

Input  Project  Scale  Scope  Speed  Budget  Portfolio  Knowledge spillovers Throughput

Internal  Project management  Innovation

management  Management/cognitive capabilities

(22)

- 22 -

Figure II.3: Overview of the dimensions of behavioural additionality

II.8 Behavioural additionality captured in routines

The previous part has shown that the concept of behavioural additionality is hard to grasp. Yet, there is a concept by which more meaning and measurability can be given to the concept of behavioural additionality. Gök (2010) suggests that the impact of behavioural additionality can be indicated by looking at the change in organisational routines. One of the most cited authors linking organizational behaviour to the use of routines, are Nelson & Winter (1982). They introduced a concept of routine-based firm evolution, in which routines are considered to act as the ‘genes’ of firm behaviour. These authors refer to routines as ‘predictable behavioural patterns’. Becker (2004) prefers to speak of routines as ‘recurrent interaction patterns’. He characterizes routines by referring to the fact that routines are regular and repetitive; that they represent an organisational activity; that they are not necessarily mindless activities; that routines are processual in nature and lastly that they are context-dependent (Becker, 2004). Furthermore, Becker (2004) indicates that organizational routines can have six identified effects. First of all they provide control and coordination. Secondly, they provide for a ‘truce’ between the person giving orders in the organization and the one executing those orders. This means that routines lead to a level of agreement in the organization about what is normal, without constantly questioning authority. Thirdly, routines ‘economize on the limited information processing and cognitive capabilities of agents’ (Becker, 2004). This means that by installing routines agents can operate without having to focus their limited attention on everything, but rather can focus on the non-routine events for which more cognitive capacity is necessary. Other effects are that non-routines reduce uncertainty and thereby provide more stability in the organization. And lastly they store knowledge in the organization (Becker, 2004).

(23)

- 23 -

occurring change. And lastly, perhaps the most interesting in this perspective, routines can help to identify pathways of exogenous drivers of change (Becker et al. 2005).

Regarding the above it can be expected that routines can be used to detect and map change in an organization. The question remains however, how to detect and classify changes in these routines. Miner et al. (2008) have used the variation-selection-retention framework, which stems from evolutionary thinking, in order to describe the relationship between organisational learning and organisational routines. By means of this framework they categorized changes in routines into three groups. First of all, there is variation in routines, which in turn exists out of two additional forms. The first form of variation is where completely new routines are created. Linked to the concept of behavioural additionality, this could mean that because the government grant enforces new rules and regulations on the recipient firm, the behaviour of that firm is likely to change. Some of these behaviours gain persistency and become recurrent patterns, therefore resulting in new routines (Gök, 2010). The second form of variation in routines results from the reconfiguration of existing organisational routines. Public policy can influence existing routines by changing their sub-routines (Gök, 2010). Public support can also lead to selection of certain organizational routines. The outcomes of a project could determine the future repetition or abolishment of routines in practice. If the public support plays a critical role in the success of a project, the routines in that particular project receive a positive selection bias and may be selected in future endeavours as well (Gök, 2010). Lastly, public support can result in the retention of organisational routines. This means that routines that are in the process of decay can be retained by the encouragement of government support. By using this characterization of routines, the evolutionary process of routines, and thus organizational change, can be more easily described and understood.

(24)

- 24 -

(Gök, 2010). This line of thought is developed by Gök (2010) in order to grasp the change in behaviour while not directly referring to the term ‘routines’. This is done because it is hard to grasp for the interviewee to define routines and may not experience their behaviour as to be captured in such.

Figure II.4: Case study approach of changing routines (Gök, 2010)

(25)

- 25 -

such as machinery, computers or even the physical lay-out (Gök, 2010). Also, this could be abstract artefacts in the form of standard operating procedures or rules and regulations. Furthermore, Gök (2010) describes the consequences of routines. These effects were already elaborated upon by referring to the work of Becker (2004). Gök (2010) has categorized them into three groups: the ‘learning/stability’ group, the ‘coordination and control/truce’ group and the ‘economizing cognitive resources/reducing uncertainty’ group. This particular categorization of aspects and consequences of routines will also be used in this case study with regard to the routine section, as it makes it easier to reach some more explicit concluding remarks on the change of routines and their effect on the organization.

(26)

- 26 -

III » Methodology

This research attempts to detect and explain additional effects that may occur at research driven companies involved in a particular subsidy program. This concerns effects additional to an increase in financial resources dedicated to research or an increase in output resulting from research. The research is constructed around several in-depth semi-structured interviews with companies involved in a subsidy program. The research is designed as a multiple case-study and takes the Innovatie Krediet as program of interest. In this section the chosen research methods will be elaborated upon and the research design will be explained in more detail.

III.1 Explanation of research methods

This research is mainly based upon the direct input derived from the participating companies in the Innovatie Krediet program. Typical R&D evaluation methods are based upon questionnaire approaches. In such an approach the data collection takes place by setting up a survey or a similar system which asks a large amount of participants to indicate their realised achievements. These questionnaires tend to have a quantitative character and are thereby suitable for discussing what happened and how often (Cooper & Schindler, 2006). From the literature however, there are some critical remarks that can be placed with this type of approach. Eisner (1998) for example, criticises ‘the technological scientism’ of experimental and quasi-experimental designs by claiming that ‘the quantitative nature of these approaches miss an important deal of issues that can only be captured qualitatively’. Georghiou (2007) states that present research techniques, and in particularly surveys, focus too much on linear and predictable effects, while missing other effects which may deem to be of even greater value. This proposition is based upon several limitations that arise from using just questionnaires in order to obtain relevant data concerning public policy effects on R&D activity. One of the limitations which are mentioned by Georghiou (2007) is that there tends to be a built-in emphasis on some expected results rather than concentrating on the unexpected. Also, inaccurate filled in surveys due to confusion or misinterpretation, may result in biased results.

(27)

- 27 -

iceberg of which much of its mass exists underneath the obvious line of sight, and thus remains hidden to the superficial observer. In order to overcome this inability to detect hidden effects, a new approach concerning the impact assessment of public funding programs has been proposed (Georghiou, 2007). This new approach puts the emphasis on detailed case studies in order to grasp the effects which lie beneath the surface. By making use of detailed case studies, observations can be put into context, interactions between observed indicators or effects can be taken into account and effects can be illustrated which could not have been detected by means of a questionnaire. While quantitative studies are providing a high level of generalizability and breadth, which may be lacking in qualitative studies, this lack is compensated for by a greater depth and detail (Sommer & Sommer, 1997).

While in most studies a group’ is implemented, this study does not make use of a ‘control-group’ for the following reasons. First of all, finding a comparable control group would prove to be quite difficult. In order to preserve a certain level of validity, this control group would have to act like an identical ‘twin-firm’. The businesses participating in the Innovatie Krediet program are operating in diverse technological niche markets. These niche markets are not likely to have many players, let alone have players with the same characteristics as the firms under investigation. So even if a ‘twin-firm’ could be selected, the difference in firm characteristics and the context in which they are operating could prove to be problematic. Drawing conclusions based on a bad comparison would not be appropriate and is therefore well-considered and deliberately left aside.

Taking in consideration the theoretical views on a qualitative versus a quantitative research approach, the following part describes the researcher’s rationale for applying a more qualitative research design in the form of in-depth interviews. First of all, the participants of the Innovatie Krediet may be

reluctant to share sensitive information about their involvement in the Innovatie Krediet program. To get a required amount of relevant information and to win the trust of the participants in order to share this information, it is critical to create a personal setting in which both researcher and interviewee can interact. Questioning the participants about their (organizational) behavior, perhaps discussing unwanted or less appropriate behavior, may stress the ease of the conversation. Participants may be subjected to questions they rather not answer in detail with an ‘unknown’ person. The development of trust in the conversation is therefore an essential part of the interviews. This is an aspect which is not likely to be achieved with a questionnaire. Also, conducting an interview may allow the interviewer to work towards answering a question in a more subtle way. Too directly asked questions, as often used in questionnaires, may scare off participants from answering. Being able to smooth the way of

(28)

- 28 -

two-sided, the interview will most likely result in bringing up more (relevant) information. Every participant has its own story, which should be considered as such. Sending out a questionnaire will therefore probably result in a too narrow view of the case in question, because mainly the view of the researcher will be taken into account and there is not much room for extra input from the participant’s perspective, while this should be the case. Another advantage of making use of an interview method is the fact that the interviewee is not bounded to a restricted set of answers. The interviewee can answer ‘freely’, which may lead to results of which the interviewers could not have thought of themselves. Especially with a research topic like behavioural additionality which is considered to be a ‘black-box’ in the innovation subsidy literature, these types of answers could lead to a better understanding of the concept.

The discussion above provides the reasoning behind the choice for a qualitative in-depth case study approach for this research. The case studies will be performed by making use of semi-structured interviews. The objective of these interviews is to try to achieve a deeper understanding of the underlying factors influencing the phenomenon of behavioural additionality and to detect different dimensions of this form of additionality. Semi-structured interviews are considered to be quite suitable for this purpose, because these are an efficient way to gather data about concepts that are not easily observed (Harrell & Bradley, 2009). Furthermore, semi-structured interviews leave room for further clarification of particular complicated subjects that in a structured interview may not be fully understood by the interviewee or may be totally differently interpreted.

After conducting the interviews, they will be transcribed literally in order to preserve a required level of detail. After the transcriptions are produced they will be analysed and coded in order to categorize the level of change and the affected dimension and routines. After this process a thorough case description will be developed of each case in which the different dimensions of behavioural additionality and the effect on routines will be described. Based on these case descriptions a more general analysis of the data will be produced in order to come to some concluding remarks.

III.2 Case Selection

(29)

- 29 -

businesses involved in the program are active in different sectors of the market, thus creating a more diverse research field. Also, the fact that not just one industry or sector is selected as topic of interest may avoid possible biases related to market specific issues. The participating businesses will be selected through consulting the annual report of the Innovatie Krediet for the period of 2009, in which all participating companies of that particular year are listed. Because the focus of this research is on behavioural changes and these changes are most likely to occur over longer periods of time, the sample will exist of some of the earliest participants in the program. The choice for the selection of participants from the year 2009 is also influenced by the fact that the executed projects take up approximately 3 years to finish (a maximum project trajectory of 4 years for the participants of the Innovatie Krediet is allowed, however the interview with the manager from the Innovatie Krediet indicated 3 years as the mean project time). With this in mind, the selected cases would all have finished or have nearly finished their project by the time of the interview. This would ensure a proper recollection of memory with regard to the project from the side of the interviewee. Also, the fact that the businesses had to comply with the SME criteria has been taken into account with the case selection. The interviews will take place with senior members of the organization who have been present in the organization at least one year prior to the participation in the Innovatie Krediet program. No cases will be selected in which a merger or acquisition has taken place after the executed project. Such a development within a case could distort the view on organisational changes and routines due to the involvement of the Innovatie Krediet and those cases are therefore deliberately left aside.

Furthermore, an in-depth interview will be conducted with an executive member of Agentschap NL. This will be done prior to the interviews with the companies in order to discuss possible high potential cases and to reflect on their view of the program’s contribution to possible behavioural additionalities. It would be interesting to see if and which discrepancies may exist between the views on behavioural additionality of both the issuer and the recipient of the program. This interview is mainly used as additional information on the program for the researcher and will therefore not be transcribed into the appendix.

III.3 Case descriptions

(30)

- 30 -

starting companies and two established firms, ranging from over thirty years of existence to just six years of existence. The six businesses were selected from a total of twenty-three possible cases; this represents approximately a quarter of all the participants involved in the Innovatie Krediet for the year 2009.

(31)

- 31 -

III.3.1 Case A: The Case of Qiy

Description of the company

Name of the company: Qiy B.V.

Founder: Marcel van Galen

Number of fulltime employees: 12

Located at: Boxtel, the Netherlands

Sector: ICT

Type of business: Starter, founded in 2006

Qiy B.V. is a company that has developed a digital personal platform, by which users can make use of all their digital data and the combination between them. Nowadays, personal digital information is stored behind several dozens of internet portals and nobody knows what exactly is stored there and who has access to this. Qiy tries to give the end-consumer back their digital identity. By creating a digital dashboard in which the consumer has an overview of all his relevant data, the consumer can decide whether or not to share data with third-parties, whether or not to combine different kinds of data, in order to restore control and privacy. Think for example of insurance, salary, medical and financial information.

Description of the interviewee

Name: Mr. Maarten Louman

Position at Qiy B.V.: Partner, responsible for R&D and Business Development

within Qiy B.V.

(32)

- 32 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling Description of the funded project

Credit received: 3.200.728 Euros

Years projected: 3 years

Stage of the project: In its final stages

This business was designed around the assessed project. It is fair to say that the business exists solely out of the activities performed within this project. Therefore the overall business case could be considered to be quite similar to the project case. This project was started to design a consumer-based platform in which important digital personal data and information could be shared, collected and controlled, in order to create a better overview and more insights. The initial idea for this project was formed in the years 2005 and 2006, from the feeling that personal digital data was freely flowing around the World Wide Web, without the required level of control and privacy that would be considered suitable for this kind of information, and the abundance of user unfriendly web-portals that control this data. This feeling resulted in the creation of a secure single platform that is created around the consumer and which makes the use of different web-portals obsolete. In those first three years the foundation for this platform was developed. Then, in the beginning of the year 2009, they ran out of money, so action had to be undertaken. A search for external financing was set up. Two informal investors saw a great business opportunity in Qiy, and decided to fund the necessary twelve million euros. This being private investors, looking for an opportunity to make money, and therefore eventually willing to sell the business, the management of Qiy decided not to take the offer of the investors, and to look for financing elsewhere.

They argued that in order to protect the data of their customers they should remain independent and were reliant on a great deal of continuity. In order to secure this level of continuity they decided to make Qiy B.V. the property of an independent non-profit foundation, the Qiy Foundation.

This solved one problem, but the financing was still an issue. The Innovatie Krediet offered them the perfect solution. They could receive a substantial amount of funding without losing their independence. So, they applied for the credit and received it in May 2009. Furthermore they made use of the Innovatief Borgstellingskrediet by which they financed their go-to-market a few years later. This can be considered to be a loan from the bank for which Agentschap NL takes eighty per cent of the guarantee. This was an amount of approximately two and a half million euros. In order to come up with the rest of the financing they started to issue profit-sharing bonds.

(33)

- 33 -

The project focused on developing the first applications, creating a demo-version of the Qiy dashboard and to develop the security technology in order to protect the data of their customers. They also started to cooperate with a Dutch store chain (Blokker), in order to arrange a salary application on Qiy for their employees. After that, several other parties were also involved in the project in order to look for opportunities. Several departments of the Dutch government were involved, such as the Dutch Tax Authorities, the Chamber of Commerce and the executive department of education (DUO). Furthermore, commercial parties such as Ziggo, Holland Casino and Post NL also participated in order to help clarify the business case of Qiy. These parties deliver their own applications to Qiy, and Qiy arranges the dataflow between different applications and the required level of security. The idea behind the dashboard is that customers can fill in information in these applications in order to give them an overview of their financial, medical or other personal situation, without this information freely flowing to the third parties without permission. So the user itself can decide whether or not to share certain information with another party. Think for example of the situation in which a user wants to receive information about a mortgage. He can let the applications of the bank, and his salary administration make an estimation of a possible mortgage without directly sharing information with these parties. This is possible because the third party, in this case a bank, has supplied the information by which the application can estimate what kind of mortgage could be obtained with a given amount of salary etcetera. This can be done ‘inside’ the application that the customer itself controls, so there is no direct information being sent to that bank. For the third parties who are participating, Qiy offers them a cheaper way to interact with their customers (decrease in the cost of sending mail) and it offers them a new way to interact with their customers.

(34)

- 34 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling Score on the dimensions

When zooming in on the project level, R&D level and the organisational level, the following results show up:

Project level

 The question whether or not the project would have been undertaken without the Innovatie Krediet could be answered with a definite ‘no’. In order to stay independent, funding from the government was inevitable and necessary. The funding did not result in performing the project on a larger scale. The funding was designed around the project; the Innovatie Krediet did not enlarge the budget. The project has also not been adjusted in terms of scope due to the funding from the Innovatie Krediet. The funding did however speed up the project. Mainly due to the involvement of Agentschap NL, third parties were found more easily and the distribution of the profit-sharing bonds was stimulated among investors.

 The processes in the project itself were also affected by the involvement of the Innovatie Krediet. With respect to the initial business plan that Qiy. B.v. presented to Agentschap NL, no major changes had been implemented. The advisors at Agentschap NL did however prove to be important sparring-partners with respect to further developing the project. Other parties that have been involved have already been discussed. These exist out of several commercial parties and public parties. The contribution of these parties was the possibility for Qiy to show its commercial potential, and to provide broad support for this new online movement.

“They were able to place themselves in the shoes of the management and for example actively thought of ways to issue new bonds. That is what Agentschap NL does better than a regular bank for instance.”

(35)

- 35 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling R&D level

 No structural budget changes have been made due to the involvement of the Innovatie Krediet. Mainly because the project is the entire business, and the entire budget is already spent on R&D. The product portfolio has also not been subjected to any changes, as there is only one product in the entire company and because there was no pre-existing portfolio.

 No additional time is being spent on innovative activities, as all time is already spent on the innovation. There were some signs that due to the intervention of the Innovatie Krediet a more structured way of working has been developed. As Mr. Louman put it:

“Yes, we are working more structurally. In order to receive the credit, a solid business case had to be presented. That has helped us to think about this more structurally, to actually make a 3-year plan.”

The case of intellectual property rights within Qiy is limited by the very nature of the business. Of course they try to protect certain aspects of the business, but their ultimate goal is to create a new open source standard, comparable with the World Wide Web. They did set up Qiy IP B.V.; as the part of the business in which their ideas are described in detail and registered. On R&D level there are some ties with businesses from the United States, these could be considered to be developing partners. There are no signs however that the Innovatie Krediet has contributed to the contact between these parties.

 The internal know-how had increased on a large scale. The question is whether that is due to the specific intervention of Agentschap NL and the Innovatie Krediet. If another party had financed the project, they would have probably reached the same increase of internal know-how. There has been a change in the innovation strategy after the project. At the beginning of the project they had decided to develop all the different applications for Qiy as well. Along the way they saw that such a thing was not possible so they decided to create a software development kit by which third parties can create their own applications. In that way they can keep the focus on developing Qiy, instead of producing applications.

(36)

- 36 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling Organizational level

 Knowledge spill-overs have probably taken place. All the third parties involved have been witnessing the developments of this new digital movement. It is therefore not unlikely that those businesses or organizations will redesign their way of interaction with their customers. This could result in new customer communication portals for example or a change in customer relationship management.

 Developments in the capabilities of the management could be described as more structurally producing their financial reports. They have hired an accountant to take care of this, where they used to do the financial reporting themselves. No specific changes in interaction with respect to their network were mentioned. The intervention of the Innovatie Krediet especially acted as a quality stamp for the business of Qiy. Investors were attracted by this, and also new business partners. Mr. Louman described this as follows:

“In the beginning, the involvement of Agentschap NL was really crucial; it was the foundation on which we were building.”

 No change in business culture was mentioned. This could very well be due to the fact that the project was started at the same time the business was founded. So, there was no pre-existing culture to be altered.

Impact on the dimensions of behavioural additionality

Project R&D Organisation

Project High

Scale Low

Scope Low

Speed High

Project management Medium

Cooperation Medium

Training None

Product/Process High

Budget Low

Portfolio Low

Innovation management Medium

Alliances Low

Know-how Medium

Innovation strategy Medium

Knowledge spillovers High

Management/cognitive capabilities High

Networking Low

Quality High

Culture None

(37)

- 37 -

University of Groningen; Faculty of Economics and Business / Master’s Thesis J.M. Tammeling Score on the routines

Routines were difficult to assess because it is a quite young business, and the project, in essence, comprehends the entire business. The only routines that had visibly changed were adopting a financial expert for the financial reporting and the more structural project planning.

Routine A-1 Hiring financial expertise:

The management of Qiy used to produce their financial reports themselves, but because of the growing complexity and emphasis on producing these reports, a financial expert was hired to take care of this. The fact that the management of Qiy realised the growing importance of professionalizing their financial reports could be considered to be the ‘ostensive’ aspect of this routine. The actual hiring of a financial expert is the ‘performative’ aspect, while the financial expert himself could be considered to be the ‘artefact’. Consequently, this leads to a reduction of time and effort the management has to put in producing these reports themselves while the level of quality of these reports is maintained. Therefore, this could be considered as ‘economizing cognitive resources/reducing uncertainty’. Routine A-2  Structuring project planning:

Through the involvement of the Innovatie Krediet the management of Qiy was triggered to develop a more structured 3-year plan for their project. They were triggered to think further ahead (ostensive aspect). This has led to more stability during the project.

Conclusion with regard to Behavioural Additionality

(38)

- 38 -

(39)

- 39 -

III.3.2 Case B: The Case of Mapper

Description of the company

Name of the company: Mapper Lithography Holding B.V.

Founders: Marco Wieland, Bert-Jan Kampherbeek & Pieter Kruit

Number of fulltime employees: 200

Located at: Delft, the Netherlands

Sector: Machinery

Type of business: Starter, founded in 2000

Mapper Lithography is a company which is currently developing a lithography machine for the chip-industry. The machine they are developing is unique in its kind because it can produce chips without ‘a mask’ and with electron beams. This reduces costs and improves performance. The company is located in Delft and has about 200 employees of which 180 work at the R&D department. The idea for developing this machine was the result of the graduation of two students of the Technical University of Delft and their supervisor. After researching the possibilities of the current Mapper technology, they were so convinced it would work, they founded Mapper Lithography. Mapper Lithography is active in the same industry as the Dutch ASML, and the Japanese multinationals Canon and Nikon.

Description of the interviewee

Name: Mr. Marco Wieland

Stimulating Innovation in the Netherlands through Governmental Interference