Master thesis Technology Management
Stimulation of innovation by the Dutch government
during an economical crisis
A comparative analysis conducted to investigate the effects of an increase of a major Dutch policy instrument, the WBSO-grant
Student
Robbert Grijseels S1329812
Rijksuniversiteit Groningen
Faculty of Economics and Business MSc. Technology Management University Supervisor: Dr. G.C. Ruël University Co-Assessor: Drs. B.N. Petkova
ABSTRACT
In this research a specific part of the relation between policy and innovation is investigated. Innovation, and in particular technological innovation, is important for both companies and countries as a whole, as new or improved products and services, and new and improved ways of production, have been for a long time the main motor of economical growth (Schumpeter – 1934 & OECD – 2006). According to Brouwer (2007) productivity is positively supported by both competition and innovation, which are fields of interest with respect to governmental policy. In 2000 the European Union agreed on a new continental policy to improve productivity in Europe. As a result various grants are applied by the EU but also by individual countries to stimulate and support (technological) innovation. In the Netherlands the largest innovation grant is the ‘Wet Bevordering van Speur- & Ontwikkelingswerk’ (WBSO-grant). This grant scheme, what has been effectuated since 1994, aims at stimulating the Research and Development (R&D) activities of Dutch companies. In 2008 were 80,030 R&D projects rewarded with WBSO-grants (Ministerie van Economische Zaken – 2009). In brief, the WBSO-grant subsidies labour costs associated with R&D activities. The more labour hours are spend on R&D, the more grant is rewarded.
As a reaction to the current financial and economical crisis the Dutch government decided to increase the contribution that companies receive out of the WBSO-grant from 01-01-2009 until 31-12-2011. With this temporarily expansion the Dutch government tried to make it more attractive for Dutch companies to invest in research & development. In the view of the endogenous growth theory (see e.g. Aghion and Howitt (1999)), competition and innovation are interrelated and as such seen as important determinants for productivity and in that way contributing to sustained economic growth. But contrarily according to Brouwer and Van der Wiel (2010) an inverted U-curve between competition and innovation exists and therefore the increase of the WBSO-grant contribution could probably lead to an opposed effect. In order to make sure that when the Netherlands are facing a new economical crisis a well founded decision about a potential increase of grants can be made; this research is focused on the relation between an increase of WBSO-grants and the productivity of Dutch companies. Therefore the impacts of the WBSO-grant were investigated at a company level. These impacts deserve further attention and research according to Jong, de & Verhoeven (2007) and Cornet (2001).
PREFACE
This Master Thesis forms the final assessment in the graduation procedure for the master degree program of Technology Management at the University of Groningen. It is the result of a research project that has started in October 2011. During a period of about 10 months, I executed a research project on the effects of innovation policy by the Dutch Government.
Performing such an academic research was a great experience for me. Besides challenging my academic competences through my research project, I have also been able to learn how to adopt an attitude resulting in tangible benefits knowledge base for both the University as the company I worked for.
I sincerely want to thank dr. Gwenny Ruël for giving me support and providing me with relevant input for this research project. As supervisor of the University of Groningen she represented the academic side of this project greatly. Her supervision and advice helped me expressively in ensuring that the empirical research had a secure theoretical basis. Moreover, I want to express my sincere gratitude to drs. Boyana Petkova for providing me input to my research project as the (independent) academic co-assessor.
Finally a note of thanks to my girlfriend, my family, and dearest friends who supported me throughout this project.
Enjoy reading this report.
Robbert Grijseels
TABLE OF CONTENTS 1. INTRODUCTION ... 7 1.1 Research objective ... 8 1.2 Problem definition ... 8 1.3 Sub-questions ... 8 1.4 Conceptual model ... 8 2. LITERATURE OVERVIEW ... 10
2.1 The WBSO-grant introduced ... 10
2.1.1 Increase of WBSO-grant during economical crisis ... 11
2.2 Innovation at a company level ... 14
2.2.1 Key indicators of innovation and how to measure those ... 15
2.3 Performance / competitiveness at a company level... 17
2.4 Relationship between amount of innovation and competitiveness of Dutch SMEs ... 18
2.5 Summary of theoretical research and hypotheses... 20
3. METHOD SECTION ... 23
3.1 Operationalised concepts ... 23
3.2 Data collection ... 24
3.3 Methods of analysis ... 25
4. RESULTS SECTION ... 28
4.1 Evidence of relation between increase of the WBSO-grant and amount innovation ... 28
4.2 Strength of relation between increase of the WBSO-grant and amount of innovation ... 39
4.2.1 Implications of WBSO-grant increase ... 39
4.2.2 Strength of relationship with respect to R&D hours ... 40
4.2.3 Strength of relationship with respect to R&D investments ... 44
5. CONCLUSION AND DISCUSSION ... 48
5.1 Conclusion ... 48
5.2 Theoretical discussion and implications ... 50
5.3 Methodological discussion and limitations ... 51
5.4 Practical implications ... 51
5.5 Future directions of research ... 51
REFERENCES ... 53
APPENDIX A: Additional tables ... 57
APPENDIX B: Multiple regression analyses ... 61
APPENDIX C: Elaborated literature overview innovation vs. competitiveness ... 66
LIST WITH ABBREVIATIONS
CBS Centraal Bureau voor de Statistiek (Dutch)
CPB Centraal Plan Bureau (Dutch)
EPO European Patent Office
EU European Union
GDP Gross Domestic Product
OECD Organisation for Economic Cooperation and Development
R&D Research and Development
SME Small and Medium Enterprises (less then 250 employees)
1. INTRODUCTION
Innovation, and in particular technological innovation, is important for both companies and countries as a whole. The development of new or improved products and services, and the development of new and improved ways of production, have been the main motor of economical growth for a long time (Schumpeter - 1934 & OECD - 2006). Although we know a lot about the positive effects of innovation, the methods to stimulate innovation are still uncertain (Cornet - 2005). This research investigates the relationship between policy and innovation at a company level.
A commonly used approach for governments to stimulate innovation is to provide a grant. In the Netherlands, the largest innovation grant is the ‘Wet Bevordering van Speur- & Ontwikkelingswerk’ (WBSO-grant). This grant scheme, what has been effectuated since 1994, aims at stimulating the Research and Development (R&D) activities of Dutch companies, which expectably lead to innovation and improved competitiveness. In 2010 19.450 companies received a WBSO-statement, e.g. a positive examination of their WBSO-application. In brief, the WBSO-grant subsidies labour costs associated with R&D activities. The more labour hours are spend on R&D, the more grant is received. In 2010 Dutch SMEs received together 860 million Euro WBSO-grants from the Dutch Government (Agentschap NL, 2011).
The foster of economic growth is especially relevant to Governments during an economical crisis. As a reaction to the current financial and economical crisis the Dutch government decided to increase the contribution that companies receive out of the WBSO-grant, from 1st of January 2009 until 31st of December 2011. With this temporarily increase the Dutch government tried to make it in particular more attractive for Dutch Small and Medium Enterprises (SMEs - less then 250 employees) to invest in R&D. In total the Dutch government increased the WBSO-grant with 150 million Euro a year (in 2009, 2010, and 2011).
However the current knowledge about the effects and the effectiveness of an increase of innovation policy, and in particular an increase of an innovation grant during an economical crisis, is very limited. Before the increase of the WBSO-grant no tangible evidence was available whether a causal relationship exists between an increase of an innovation grant and the amount of innovation by receiving companies. As a result the decision to increase the WBSO-grant could be classified as experimental policy.
to policy makers when a new economical crisis appears. These outcomes deserve further attention and research according to Jong, de & Verhoeven (2007) and Cornet (2001).
The remaining of this chapter will firstly describe the purpose of this research in the Research objective (1.1). Thereafter the problem definition, verified in this thesis, will be stated (1.2), with the corresponding sub-questions (1.3). And finally, in 1.4 an overview of the different concepts of this research will be given through the conceptual model.
1.1 Research objective
This research aims at investigating the relationship between policy that endeavours to stimulate company innovation and the amount of innovation by Dutch companies. This is done by retrieving relevant insights about the impacts of an increase of the Dutch WBSO- grants during the first phase (2007 - 2010) of the current financial and economical crisis at a company (SME) level. These insights should be based on both theoretical and empirical evidence and should result in more detailed information about the relationship between policy and innovation.
1.2 Problem definition
This research is shaped by the following problem definition:
To what extent had the increase of the WBSO-grants, by the Dutch government during the first phase of economic crisis (2007-2010), led to more innovation by and better performance / competiveness of Dutch SMEs that received this additional support?
1.3 Sub-questions
To answer the problem definition the following sub-questions will be handled:
a. What is the WBSO-grant (1) and in what way is the WBSO-grant increased (2)?
b. What is innovation (1), what are the key indicators of innovation (2) and how can innovation and its key indicators be measured (3)?
c. What is performance / competitiveness at a company level?
d. In what way is the amount of innovation by Dutch SMEs related to the performance / competitiveness of these SMEs, according to literature?
e. In what way is the increase of the WBSO-grant related to the amount of innovation by Dutch SMEs (1)? How strong is this relationship (2)?
1.4 Conceptual model
Figure 1: Conceptual model
Based on relationships competition, innovation, productivity, and policy (Brouwer - 2007)
The amount of additional received WBSO-grant is handled as the independent variable within this research. The performance / competitiveness of Dutch SMEs is the dependent variable with the amount of innovation as a moderator.
Within this research a literature exploration of the theoretical impacts of innovation grant schemes, and an empirical research of Dutch SMEs that receive WBSO-grants, was executed both. For retrieving empirical evidence, a set of companies (N = 122) that received the increased WBSO-grant in 2009, 2010 and 2011 was analysed with Paired-Samples T-Tests, Pearson correlation analyses and regression analyses.
The next section (chapter 2) presents the outcome of a literature exploration. Chapter 3 handles the research methodology and measurement instrument validation and chapter 4 offers the results of the empirical research. The final section (chapter 5) of this work presents a conclusion, a discussion of the results, directions for further research and limitations of this study.
2. LITERATURE OVERVIEW
In this chapter a summary of an extended literature research is presented, in line with the, in the previous chapter presented, problem definition and related research questions. This chapter contains the following subsections:
2.1 The WBSO-grant introduced
2.1.1 Increase of WBSO-grant during economical crisis 2.2 Innovation at a company level
2.2.1 Key indicators of innovation and how to measure those 2.3 Performance / competitiveness at a company level
2.4 Relationship between amount of innovation and competitiveness of Dutch SMEs 2.5 Summary of theoretical research and hypotheses
2.1 The WBSO-grant introduced
In this section a short introduction of the WBSO-grant is presented.
The WBSO-grant (Wet Vermindering Afdracht Loonbelasting en Premie voor de Volksverzekeringen, Speur- en Ontwikkelingswerk), is a tax facility with which the Dutch government aims to stimulate innovation and, in particular, Research & Development. Hereto, the WBSO-grant offers the opportunity to deduct a substantial part of the labour costs of own personnel that is directly involved in R&D activities, from the labour tax a Dutch company has to pay.
In 2008 the WBSO-grant amounted 42% per company over the first 110.000 Euro cumulated gross labour costs and 14% over the remaining accumulated gross labour costs. For companies younger than five years, and that have not yet made use of the WBSO-grant thrice, the percentage over the first 110.000 Euro will even amount to 60%. The maximum annual WBSO-grant was 8.000.000 Euro per fiscal conglomerate. The total WBSO-grant budget was 550 million Euros in 2008. The details of the WBSO-grant vary by year. As pointed out the above numbers were valid in 2008, before the increase of the WBSO-grant. In section 2.1.1 the numbers for 2008, 2009, 2010 en 2011 are presented.
Calculation example former WBSO-grant benefits in 2008
The WBSO-grant distinguishes four different types of R&D-projects:
1. Development of technically new physical products, physical production processes, software or components thereof.
2. Technical-scientific research seeking to explain phenomena in fields such as physics, chemistry, biotechnology, production technology and information and communication technology.
3. Analysis of the technical feasibility of an R&D-project one intends to launch.
4. Technical research aimed at enhancing your physical production process or software.1
Additionally the projects are divided in R&D-activities with respect to physical products and processes and R&D-activities with respect to software. Underneath a brief description of the R&D-activities that qualify for a WBSO grant is presented.
Systematically organised activities that are executed within the European Union by employees paid by a Dutch company. Furthermore those activities should be directly and solely aimed at technical-scientific research or the development of (components of) physical products or production-processes which are technically new for the company involved. Also feasibility-studies that precede these development activities qualify for a WBSO grant.
Important in above definition is the fact that the WBSO only requires development activities to be technically new for the applying company. As such, development projects that already have been executed by other companies can still qualify for a WBSO-grant.
In relation to R&D in the software domain, the following activities qualify for a WBSO grant.
Systematically organised activities that are executed within the European Union by employees paid by a Dutch company. Furthermore those activities should be directly and solely aimed at technical-scientific research or the development of (components of) software which is technically new for the company involved. Also feasibility-studies that precede these development activities qualify for a WBSO grant.
In relation to software development projects, the WBSO-facility also searches for R&D-activities that are technically new for the company involved.
2.1.1 Increase of WBSO-grant during economical crisis
Available research on innovation suggests that an economical crisis affects the amount of innovation. Historically, business R&D-expenditure and patent filings from the European Patent Office (EPO) have moved in parallel with Gross Domestic Product (GDP) a measure for productivity, slowing markedly
1
during the economic downturns of the early 1990s and of the early 2000s (Figure 2) according to the Organisation for Economic Cooperation and Development (OECD - 2009).
Figure 2: Relation GPD, EPO patents and Industry R&D
Source: OECD (2009) annual growth rate for the total of the OECD countries divides by standard deviation.
As stated in the OECD (2009) report: “R&D is declining because it is mainly financed from cash flow (retained earnings), which contracts in downturns. At the same time, as banks, markets and investors have become more risk averse; firms face difficulties in tapping into external sources of funding to support their investments in R&D. Business R&D is also being re-oriented towards short-term, low-risk innovations, while longer term, high risk innovation projects are being cut first. The decline in business R&D risks affecting the stock of knowledge as highly trained researchers and innovators lose their jobs. Small, innovative firms are particularly hard hit because in many cases their primary asset is intangible in nature (e.g. an idea or a patent) and difficult to value, making it hard to borrow against, or sell, to stay afloat.”
Therefore as a reaction to the current financial and economical crisis and to cope with above dynamics the Dutch government decided in 2009 to increase the contribution that companies receive out of the WBSO-grant from the 1st of January 2009 until the 31st of December 2011. The Dutch Government made an additional 150 million Euro a year available for the WBSO-grant. With this extra money the tax benefit through the WBSO-grant increased: companies needed to pay less income tax. Hereby the Dutch government tried to make it more attractive for Dutch companies to invest in R&D because labour costs for innovation were reduced.
In 2009, the increase of the WBSO-grant consisted of the following:
• the rate of the first disc increased from 42% to 50%. • the percentage of the 2nd disc increased from 14% to 18%. • the ceiling increased from 8 million Euros to 14 million Euros.
In 2010 and 2011 the increase of the WBSO-grant included the following:
• the upper limit of the first disc increased from 110.000 Euro to 220.000 Euro. • the rate of the first disc increased from 42% to 50%.
• the percentage of the 2nd disc increased from 14% to 18%. • the ceiling increased from 8 million Euros to 14 million Euros.
• for companies younger than 5 years, and that have not yet made use of the facility WBSO thrice, the grant rate over the first 220.000 Euro here amounted to 64%.
Based on this background of the WBSO-gran we formulated the following hypothesis:
H1. The increase of the WBSO-grant resulted in more R&D activities by Dutch SMEs during the years of the increase.
In the frame below we present a calculation example of impacts of the increased WBSO-grant at a company level.
Calculation example increased WBSO-grant benefits
Company A has 100 employees on the payroll and the total of labour tax is 1.000.000 Euro a year. 15 of the employees are involved with the R&D activities of Company A and on average they spend 400 hours a year on R&D. With an average gross hour wage of 20 Euro the sum of the yearly R&D wage cost turns out to be (15 * 400 * 20 =) 120.000 Euro. The WBSO tax deduction will be 50% of 220.000 Euro. In total Company A will receive a total of 60.000 Euro WBSO-grant a year.
By this way of increasing the WBSO-grant the Dutch government stimulated small companies more then larger companies (Agentschap NL, 2011). Based on this theoretical discovery we formulated the following hypothesis:
H2. The increase of the WBSO-grant resulted in more R&D activities of small Dutch companies in contrary to medium and large companies during the years of the increase.
innovation more severe then capital intensive innovation. Based on these theoretical findings we formulated the following hypothesis:
H3. The increase of the WBSO-grant resulted in more R&D activities of Dutch software development companies in contrary to physical development companies during the years of the increase.
In chapter 4 the effects and effectiveness of the WBSO-grant will be evaluated more extensive.
2.2 Innovation at a company level
An important feature of the WBSO-grant is to stimulate innovation at a company level. In this first part of this chapter the concept of innovation will be explored. In second part the measurement of innovation will be handled.
Organizations want to deliver value to customers and other stakeholders. It is the reason they exist. To ensure continuity, these organizations need to make sure they can keep on creating and delivering value. This can be done by improving the way they currently create value, or by finding new ways to create value. In other words: they need to innovate.
There are various definitions of innovation that appear in literature. Innovation is a comprehensive and broad conceptual understanding, which may give rise to many and varying definitions as described by Lanser & Van de Wiel (2011). The purpose of this section is to present some of the major definitions.
According tot Rogers (1998) Joseph Schumpeter is often thought of as the first economist to draw attention to the importance of innovation. He defined in the 1930s, five types of innovation:
1. the introduction of a new product or a new quality of an existing product; 2. the introduction of a new method of production;
3. the opening of a new market
4. the development of new sources of supply for raw materials or other inputs 5. changes in industrial organisation
Schumpeter (1934) explains the importance of innovation by the notion of creative destruction. He states that the value from the old is destroyed by the new. In capitalist reality, it is not price competition that counts, but the competition from a new commodity, technology, source of supply or type of organization, which strikes existing firms at their foundations. This kind of competition is much more effective than the other and is the powerful lever that in the long run expands output.
firms. Utterback (1994) describes that most industry-shattering innovations do not spring from established competitors in an industry but from new firms or from established firms entering a new area.
Hence, to find new or improved ways to create value, organizations wish and need to innovate. As described by Chesbrough (2003), it is not why to innovate, but how to innovate that remains the subject of much research and discussions. Increased competition, shorter innovation cycles, research and development’s increasing costs, as well as the scarcity of resources are some of the reasons for companies to look for new ways to innovate. Moreover Cornet & Vroomen (2005) mention innovation as an important determinant of labour productivity growth. And Crossan & Apaydin (2010) see innovation as both a process and an outcome: the production or adoption, assimilation, and exploitation of a value-added novelty in economic and social spheres; renewal and enlargement of products, services, and markets; development of new methods of production; and establishment of new management systems.
But as Evangelista & Vezzani (2010) pointed out, enlarging the analysis of innovation beyond the technological domain provides a much richer and complex picture of firm’s innovation strategies and performances. They argue that the traditional distinction (or dichotomy) between product and process innovation is not sufficient to depict the variety of innovation modes, which differentiate firms innovation strategies both within the manufacturing and service sector. Additionally Ulrich & Eppinger (2008) describe that successful product development and thereby innovation is an interdisciplinary activity requiring contributions from nearly all the functions of a firm; however, three functions are almost always central to a product development project: marketing, design and manufacturing.
In this research the later two (design and manufacturing) are investigated and related to technological innovation. The focus is led on technological innovation and therefore the definition of the new Oslo manual (3rd edition, 2005) is sufficient. According to the new Oslo manual (2005):
“Technological product and process innovation is the implementation or commercialization of a product or process with (far) better technological characteristics.”
2.2.1 Key indicators of innovation and how to measure those
In this section key indicators of innovation are described. Because innovation is the main topic of the empirical part of this research, the concept innovation will be operationalised in chapter 3.
an impact on economic behaviour as do considerations of transaction costs or economies of scale, yet less about them is known.
However according to Rogers (1998), Kleinknegt et al. (2002) and Brouwer (2007) innovation can be measured and therefore a division in input and output indicators is essential.
Input indicators
The level of research and development expenditure has been the most extensively used proxy for the level of innovative effort (input). Other input indicators of innovation are: number of R&D employees, amount of acquisition of technology from others, level of expenditure on tooling-up industrial engineering and manufacturing start-up associated with new products, and also intellectual property statistics according to Rogers (1998) and Kleinknecht (2002).
Based on this theoretical exploration we formulated the following hypothesis:
H4. The increase of the WBSO-grant resulted in additional R&D employees of Dutch SMEs during the years of the increase.
Aghion et al. (2005) applies also patent citation as an indicator for innovation. But Boone, et al. (2007), Kleinknecht et al. (2002) and Brouwer (2007) argue that R&D efforts and innovation rate as indicators of innovation are more robust.
Furthermore Braaksma & Reijaard (2007) describe that the WBSO-grant is a tax deduction facility for company R&D-wage costs and that by this the admission to the WBSO-grant can be considered a “hard” indicator for actually performing R&D.
Under the influence of endogenous growth theory in general R&D efforts and R&D investments so far are considered a good proxy for efforts in the field of innovation, according to Lanser & van de Wiel (2011).
Based on this theoretical finding we formulated the following hypothesis:
H5. The increase of the WBSO-grant resulted in additional R&D investments by Dutch SMEs during the years of the increase.
Output indicators
efficiency of the innovation process. A low R&D intensity can still lead to high competitiveness, when the innovation process is arranged very efficient.
Ultimately, the key measure of innovative activity is the success of the firm, which is an output indicator. Firm success can be approximated by profits, revenue growth, share performance, market capitalization or productivity, amongst other indicators. All of these indicators have drawbacks and importantly, can be caused by factors other than the level of innovativeness. However, the extent of firm success can be used as a measure off innovativeness if certain econometric techniques are used. Strength of such techniques is that the extent of a firm’s innovativeness can be quantified and directly computed compared to other firms. Other indicators of innovation are: number of new or improved products and intellectual property (IP) measures according to Rogers (1998) and Kleinknecht et al. (2002).
On the other hand the latter are not so useful for European innovation according to Brouwer (2007). He states that European patents are still too expensive compared to U.S. patents. The single market for patents in Europe is not functioning properly.
In line with the conceptual model of this research, competitiveness will be used as the output measure of a firm’s innovative performance. This concept will be explored further in section 2.3.
2.3 Performance / competitiveness at a company level
In this section competitiveness as possible output measurer of innovation is defined and operationalised.
Kendrick (1956) argues that competitiveness can be denoted as the ratio of output to any related input or class of inputs. In this sense, there is a spectrum of productivity ratios, each of which indicates the savings achieved in particular cost elements over time as a result of changes in productive efficiency and factor substitutions. In order to attempt to measure changes in productive efficiency as such, however, output must be related to the aggregate of corresponding inputs according to Kendrick (1956).
In order to measure competitiveness of a company, it is necessary to operationalise the same concept of productivity as in a production unit or a company, yet, the object of modelling is substantially wider and the information more aggregate according to Saari (2006).
Productivity is considered by the OECD (2008) as a key source of economic growth and competitiveness and, as such, is basic statistical information for many international comparisons and country performance assessments.
As an economic statistic, productivity is measured as value added, or the sales minus the cost of inputs. At the plant level, in addition to being kept in monetary units, input statistics are commonly kept as weights or volumes of raw or semi-finished materials, kilowatt hours of power, worker hours, etc. There are different measures of productivity and the choice between them depends either on the purpose of the productivity measurement and/or data availability.
Based on this theoretical research we formulated the following hypotheses:
H6. The increase of the WBSO-grant resulted in better performance assessments for Dutch SMEs during the years of the increase.
2.4 Relationship between amount of innovation and competitiveness of Dutch SMEs In this chapter the relationship between innovation and competitiveness will be explored.
Innovation has been identified on many occasions, as the main driver for companies to prosper, grow and sustain profitable (e.g. Drucker - 1988; Christensen - 1997; and Thomke - 2001). One could even argue that innovation is a precondition for doing business. Even as competitive advantage may come from size, or possession of assets, and so on, the pattern is increasingly coming to favour those organizations that can mobilise knowledge and technological skills and experience to create novelty in their offerings (product/service) and the ways in which they create and deliver those offerings (Kay, 1993).
Table 1: Link between Porter’s generic strategies and innovation Generic strategy
Overall cost leadership Overall differentiation Focus-segment cost leadership Focus-segment differentiation Technological policies Product technological change Product development to reduce product cost by lowering materials content, facilitating ease of manufacture, simplifying logistical requirements, etc. Product development to enhance product quality, features, deliverability, or switching costs Product development to design only enough performance for the segment’s needs Product design to meet exactly needs of particular business segment application Process technological change Learning curve process improvement, process development to enhance economies of scale Process development to support high tolerances, greater quality control, more reliable scheduling, faster response time to orders, and other dimensions that improve the ability to perform Process development to tune production and delivery system to segment needs in order to lower cost Process development to tune the production and delivery system to segment need in order to improve performance Source: Porter (1983)
As described in the table above both the development of technological new products and processes can lead to competitive advantages. Several studies of patterns of industrial innovation indicate that relationships exist between product and process changes, the state of evolution of an industry, and the competitive climate face by individual firms according to Utterback (1994).
According to Brouwer (2007) productivity is positively supported by both competition and innovation, which are fields of interest with respect to governmental policy. In 2000 the European Union (EU) agreed on a new continental policy to improve productivity in Europe2. As a result various grants are applied by the EU but also by individual countries to stimulate and support (technological) innovation.
Furthermore Brouwer (2007) states that firms innovate to raise their profits. This can happen in a number of ways. The innovation can reduce the firm’s production costs (this is usually called a process innovation) and hence may increase the firm’s profits relative to its competitors that do not innovate. A firm’s profit can also increase due to a product innovation (i.e. new good or services). In this case, the innovation differentiates a firm’s product from its competitors’ products. This differentiation can be horizontal (preferred by some consumers but not by all) or vertical (preferred by all consumers if sold at the same price). With such product innovation, the firm is able to raise its product margin for two reasons. One is that consumers like the product better and hence are willing to pay more. The other
2
reason is that by moving away from its competitors the firm gains competitive advantage (competitive pressure is reduced) which allows it to raise its price.
For most researchers, innovation outcome is the endpoint of their expedition. However, linking innovation outcomes with performance is critical in addressing whether and how innovation creates value. According to Mone et al. (1998), innovation capability is the most important determinant of firm performance. This theoretical suggestion has been empirically supported by several studies, according to Crossan & Apaydin (2010) and Evangelista & Vezzani (2010). In Appendix C we present an elaborated overview of studies with respect to innovation and competitiveness.
Both innovation and competition are regarded in literature as the driving forces behind productivity growth (OECD, 2001). About the relationship between competition and innovation a multitude of empirical and theoretical literature is available with varying results. Nevertheless higher productivity can be achieved by investing more in innovation and by promoting competition, but do need to be well thought out according to Boone (2000), Brouwer (2007) and Van Leeuwen (2009).
According to literature, innovation is expected to have a positive effect on the competitiveness of Dutch companies and thus SMEs.
However too high competitive advantage could have negative impact on competition, what may slow down innovation. Therefore innovation should not be intensified too much according to Brouwer (2007). This latter statement will be investigated further in chapter 4.
In the view of the endogenous growth theory (see e.g. Aghion & Howitt -1999), competition and innovation are interrelated and as such seen as important determinants for productivity and in that way contributing to sustained economic growth. But contrarily according to Brouwer & Van der Wiel (2010) an inverted U-curve between competition and innovation exists and therefore the increase of the WBSO-grant contribution could probably lead to an opposed effect.
Based on above premises the following opposing hypotheses were formulated:
H7. A linear relationship exists between the amount of additional received WBSO-grant and the number of additional R&D hours during the years of the increase.
H8. A linear relationship exists between the amount of additional received WBSO-grant and the amounts of additional R&D investments during the years of the increase.
2.5 Summary of theoretical research and hypotheses
As discussed in the prior sections, according to literature, innovation has a positive effect on the performance / competitiveness companies and thus Dutch SMEs. One remark has to be made. Too much innovation could have negative impact on competition what may slow down the performance. Therefore innovation should not be intensified too much according to Schumpeter (1942), Boone (2000) and Brouwer (2007).
The theoretical research is conceptualised in the next theoretical framework.
Figure 3: Theoretical framework
Based on relationships competition, innovation, productivity, and policy (Brouwer - 2007)
In the empirical part of this research the relationship between additional WBSO-grant (policy) and the amount of innovation by Dutch SMEs will be investigated more intensively. We summarised the concepts of the theoretical research in Table 2.
+/-Table 2: Summary of concepts
Concept Definition Chapter
WBSO-grant A tax facility with which the Dutch government aims to stimulate innovation and, in particular, Research & Development
2.1
Increase of the WBSO-grant
Additional 150 million Euro made available a year by the Dutch government
2.1
Innovation The implementation or commercialization of a product or process with (far) better technological characteristics.
2.2
Competitive advantage
The strategic advantage one business entity has over its rival entities within its competitive industry.
Appendix D
Productivity Key source of economic growth and competitiveness and, as such, is basic statistical information for many international comparisons and country performance assessments.
2.3
Relation: Innovation
Competitiveness
According to literature, innovation is expected to have a positive effect on competitiveness of companies and thus of Dutch SMEs.
2.4
We summarised the hypotheses in Table 3:
Table 3: Summary of hypotheses
H1 The increase of the WBSO-grant resulted in more R&D activities by Dutch SMEs during the years of the increase.
H2 The increase of the WBSO-grant resulted in more R&D activities of small Dutch companies in contrary to medium and large companies during the years of the increase.
H3 The increase of the WBSO-grant resulted in more R&D activities of Dutch software development companies in contrary to physical development companies during the years of the increase. H4 The increase of the WBSO-grant resulted in additional R&D employees of Dutch SMEs during
the years of the increase.
H5 The increase of the WBSO-grant resulted in additional R&D investments by Dutch SMEs during the years of the increase.
H6 The increase of the WBSO-grant resulted in better performance assessments for Dutch SMEs during the years of the increase.
H7 A linear relationship exists between the amount of additional received WBSO-grant and the number of additional R&D hours during the years of the increase.
H8 A linear relationship exists between the amount of additional received WBSO-grant and the amounts of additional R&D investments during the years of the increase.
3. METHOD SECTION
In this chapter we discuss the method of the empirical research we executed with respect to the effects of an increase of the WBSO-grant. Objective of the empirical research was to present answers to the next two sub-questions:
1. In what way is the increase of the WBSO-grant related to the amount of innovation by Dutch SMEs?
2. How strong is the relationship between the increase of the WBSO-grant and the amount of innovation by Dutch SMEs?
An answer to these questions is presented by evaluating the hypotheses, which were formulated in the previous chapter. This chapter contains the following subsections:
3.1 Operationalised concepts 3.2 Data collection
3.3 Methods of analysis
3.1 Operationalised concepts
Before the data was collected we first operationalised the different concepts of this research. Based on literature research performed in chapter 2 we distinguish 1) an independent variable, 2) a moderating variable and 3) a dependent variable.
1) The independent variable of this research, the increase of the WBSO-grant, is operationalised as the additional received WBSO-grant in 2009 in Euro.
2) The moderating variable of this research, the amount of innovation, is operationalised as the amount of R&D activities by a Dutch SME in labour hours a year. The amounts of R&D employees and the net R&D investments (with reduction of WBSO-grant) in Euro a year will be used as proxies. Furthermore we used the additional R&D hours in 2010 and 2011 as the operationalised measures for additionality, see section 3.3.
3) The dependent variable of this research, the competitiveness / performance, is operationalised as the ratio of revenues per labour cost (productivity rate). Turnover in Euro a year, profits in Euro a year and total employees a year will be used as proxies. Outcomes with respect to the dependent variable should be handled with the most care, because a lot of other variables also influence the above measures.
Table 4: Summary of operationalised concepts
Concept Definition Chapter
Increase of the WBSO-grant
Additional 150 million Euro made available in 2009, 2010, and 2011 by the Dutch government
Operationalised measure:
Additional received WBSO-grant in 2009 in Euro
2.1
Amount of
innovation by Dutch SMEs
The implementation or commercialization of a product or process with (far) better technological characteristics.
Operationalised measures:
Input indicator = amount of R&D hours a year
Proxies: number of R&D employees a year and amount of net R&D investments a year (in euro)
Additionality measure = additional R&D hours in 2010 and 2011
Output indicator, see performance / competitiveness
2.2
Performance / competitiveness of Dutch SMEs
Key source of economic growth and productivity and, as such, is basic statistical information for many international comparisons and country performance assessments.
Operationalised measure:
Ratio of revenues per labour cost (productivity rate) Proxies: turnover (in Euro) and number employees a year
2.3
3.2 Data collection
In order to test the hypotheses we collected data from 122 Dutch SMEs which have used the WBSO-grant in the period 2007 to 2012 one or more times. In Appendix A, tables with descriptive statistics about our sample are available.
In line with the operationalised measures, data about the amount of R&D3, the number of R&D employees, and company key figures such as turnover, total wages per year, and number of employees were retrieved.
The data is sorted by SME and by year. During the period of the dataset (2007 to 2012) SMEs started and ceased to use the WBSO-grant. As a result the set of SMEs in 2007 differs from the set of SMEs in 2008, and moreover the set of SMEs in 2007 differs considerably from the set of SMEs in 2012.
Furthermore some SMEs (N = 5 in 2007, N = 8 in 2008, N = 8 in 2009, N = 11 in 2010 and N = 11 in 2011) did not request the WBSO-grant every year for the full twelve months. Mostly (1) because the specific SME was not familiar yet with the WBSO-grant, sometimes (2) because the specific SME was just started, and sometimes (3) because the specific SME did no longer make use of the WBSO-grant.
3
In the first (1) case, we extrapolated the period of request towards a full year, because we argued that the specific SME already performed R&D activities but not yet within de WBSO-grant. In the second (2) and third (3) case, we did not take the figures for that year into account.
At the moment of the data collection for 13 companies the amounts of R&D hours for the full year of 2012 were available. For comparison reasons the amounts of R&D hours of the other 63 companies were all extrapolated in 2012. All these results are marked with a *.
3.3 Methods of analysis
In harmony with the constructed hypotheses we first investigated whether the increase of the WBSO-grant led to significant differences in the subsequent years. If these differences existed we tried to retrieve insight about the direction and intensity of the possible relationship.
At first we wanted to test whether or not the increase of the WBSO-grant in 2009, 2010 and 2011 influenced the amount of innovation by Dutch SMEs during those years. Normally, comparative analysis is performed with a reference group with the same specifics as the group that is subject to the change. But as pointed out by Cornet (2005) and Boone & Van der Wiel (2007) the WBSO-grant has a very high penetration rate with respect to innovative companies, e.g. the vast majority of the innovative companies utilise of the WBSO-grant. Therefore it is not possible to design a convincing reference group. Nevertheless we compared any significant outcomes with national innovation figures. By treating national figures as a kind of reference group we tried to minimise other influences, which we first neglected in line with a ceteris paribus assumption.
In order to cope with the absence of a convincing reference group and to satisfy the focus of the research, a within subject design is essential. The fact that subjects act as their own control group provides a way of reducing the amount of error arising from natural variance between characteristics of companies. As a result of the within subject design, we used years before and after the increase (2007, 2008, and 2012) as a control group with respect to years during the increase (2009, 2010, and 2011).
To test H1, H2, H3, H4, H5, and H6 we utilised Paired-Samples T-Tests. A Paired-Samples T-Test compares the means of two variables. To do so the differences between two variables for each case are computed and tested to see if the mean difference significantly differs from zero. In our research we investigated whether or not differences appeared between means of analysed years before, during, and after the increase of the WBSO-grant.
in different periods and that way we tried to minimise the influence of single and/or accidental exceptions. Furthermore this design provided additional insights in the long term effects of the increase.
Secondly, we focused on the strength of possible relationships that we retrieved out of the Paired-Samples T-Tests. According to Cornet & Vroomen (2005) it is difficult to verify whether the relationship between policy (the increase of the WBSO-grant is policy) and innovation suggests a causal relationship or not (see also Brouwer et al (2002) and Cornet (2001)). Most difficult is to distinguish substitution from additionality (Brouwer et al - 2002 and Cornet & Vroomen - 2005), which are explained to be two sides of same coin. Substitution occurs when an allocated grant is used by a company to save on their own R&D investments. By this no additional R&D activities are executed. Additionality on the contrary suggests that companies use the allocated grant to start new (additional) R&D activities.
The aim of our analysis in section 4.2 includes determining whether there is substitution or additionality and all shades in between. According to Strom (2006) such an analysis is quite difficult. Although researchers agree that government support for business R&D is beneficial for society, minor data is available about how and to what extent. Most industrialised countries use different approaches. The economic and industrial structures of countries differ, the level of R&D as percentage of GDP and the distribution of R&D between the public and private sectors and between industries varies, as do the tax systems and R&D and industrial policy frameworks. Strom (2006) concludes: “Due to these differences, there is not one single answer to whether a country should use tax incentives to increase R&D, how such a system should be designed or how it should be evaluated.”
In the literature overview (chapter 2) we hypothesised that relationships between the increase and the amount of innovation in labour hours and investments exist. To test these H7 and H8 we designed a four stage approach and by this we tried to evaluate the strength of possible relationships. First we plotted a graph. Secondly we used correlation analyses. Thirdly we used single regression analyses and fourthly we used multiple regression analyses.
For all these analyses we applied simplifying assumptions (ceteris paribus) in order to construct an analytical framework that does not necessarily prove cause and effect but still could be useful for describing fundamental concepts within the realm of this research. The frames that follow below present short descriptions of the used tests.
Correlation analysis
A correlation measure of +1 indicates that two variables are perfectly related in a positive linear sense, a correlation coefficient of -1 indicates that two variables are perfectly related in a negative linear sense, and a correlation coefficient of 0 indicates that there is no linear relationship between the two variables.
Regression analysis (single and multiple)
Furthermore with regression analysis we tried to understand whether or a significant linear model between the increase of WBSO-grant (independent variable) and the amount innovation by Dutch SMEs (dependent variable) can be constructed. By this analysis a model of the relationship is hypothesised, and estimates of the variable values are used to construct an estimated regression model. If the model has significant constants, the estimated regression model can be used to predict the value of the dependent variable given values for the independent variable(s).
Correlation and regression analysis are related in the sense that both deal with relationships among variables. Neither regression nor correlation analyses should be interpreted as establishing cause-and-effect relationships. They can only indicate how or to what extent variables are associated with each other. The correlation coefficient measures only the degree of linear association between two variables. Therefore we handled any conclusions about a cause-and-effect relationship with care.4
4
4. RESULTS SECTION
In this section we present the results of the comparative analysis we conducted. With this comparative analysis we investigated the relationship between an increase of the WBSO-grant and the amount of innovation by Dutch SMEs. We tested the hypotheses we formulated in chapter 2 one by one. In the first section (4.1) we show evidence of a relationship between the increase of the WBSO and an increase in the amount innovation. In the second section (4.2) we present approximations of the strength of the relationship.
4.1 Evidence of relation between increase of the WBSO-grant and amount innovation
First we analysed whether or not a relationship exists. In this section we present the outcomes of Paired-Samples T-Tests, which belong to H1, H2, H3, H4, H5, and H6.
As mentioned before, the increase of the WBSO-grant was initiated to stimulate innovation during the first phase of the economical crisis. Therefore we first tried to identify effects with respect to the input side of innovation. Hereto we analysed the operationalised measure, number of R&D hours, with the next hypothesis.
H1. The increase of the WBSO-grant resulted in more R&D activities by Dutch SMEs during the years of the increase.
Table 5: Number of R&D hours (all SMEs) - Paired-Samples T-Test Compared years Mean difference
(R&D hours)
Sig. (2-tailed) N
One year period
2007 & 2008 -398 ,538 37
2008 & 2009 -375 ,687 42
2009 & 2010 613 ,020 53
2010 & 2011 321 ,269 87
2011 & 2012* 247 ,202 75
Two year period
2007 & 2009 -932 ,621 31
2008 & 2010 341 ,764 40
2009 & 2011 1085 ,032 43
2010 & 2012* 504 ,093 73
Three year period
2007 & 2010 1752 ,078 28
2008 & 2011 1982 ,048 31
2009 & 2012* 1572 ,035 35
Four year period
2007 & 2011 2285 ,108 22
2008 & 2012* 2686 ,057 25
Five year period
2007 & 2012* 2942 ,149 17
* 2012 data were extrapolated
In this research we only treat mean differences with p < ,05 as significant (underlined in all tables).
The above analysis identified significant mean differences (md) between the numbers of R&D hours performed by Dutch SMEs in 2009 & 2010 (one year period, md [53] = 613, p < .020), 2009 & 2011 (two year period, md [43] = 1085, p < .032), 2008 & 2011 (three year period, md [31] = 1982, p < .048), and 2009 and 2012 (three year period, , md [35] = 1752, p < .035). In all these compared periods a significant growth in numbers of R&D hours took place.
Furthermore we state that the growth after 2009 may very well be caused by the increase of the WBSO-grant. As mentioned the WBSO-grant was increased in 2009 and again (to a small extent) in 2010 and 2011. However, the increase of 2009 was announced during that year and at that time companies could no longer make adjustments in their WBSO-request and thus to their amount of requested R&D hours. This is caused by the fact that a WBSO-grant application should be filed at least a month before the start of the development activities. In this case (amount of requested R&D hours in 2009) at least before the 1st of December 2008.
Above practical implications of the increase of the WBSO-grant explain why the numbers of requested R&D hours were not yet significantly higher in 2009 compared to 2008 and before: when companies became aware of the increase of the WBSO-grant in 2009 they could only adjust their R&D budgets for 2010 and later.
The results of Table 5 are in line with above analysis and as a result H1 is accepted. In section 4.2 we present the results of further analysis of this possible relationship. Nevertheless, we already concluded that the decision to increase the WBSO-grant during 2009 did not lead to significant differences in the amount of R&D hours in 2009. Most probably because it was not longer possible for companies to retroactively adjust their R&D budgets and initiate additional R&D hours.
We found evidence that the increase of the WBSO-grant resulted in more R&D activities by Dutch SMEs during the years of the increase. Nevertheless, the growth of additional R&D activities took place with a delay of one year. Most probably because of the choice to increase the WBSO-grant mid 2009 and at that time companies could not change their R&D budget for 2009 anymore.
To retrieve more SME specific insights we divided the sample in small (10 or less employees), medium (10-50 employees), and large SMEs (more than 50 employees). As described in section 2.1, the WBSO-grant is shaped in a way that small companies receive relatively more amount of grant in comparison to larger companies.
H2. The increase of the WBSO-grant resulted in more R&D activities of small Dutch companies in contrary to medium and large companies during the years of the increase.
Table 6: Number of R&D hours (small SMEs) - Paired-Samples T-Test
Compared years Mean difference (R&D hours) Sig. (2-tailed) N One year period
2007 & 2008 71 ,811 14
2008 & 2009 233 ,205 10
2009 & 2010 327 ,087 17
2010 & 2011 -212 ,100 39
2011 & 2012* 671 ,013 34
Two year period
2007 & 2009 374 ,215 8
2008 & 2010 638 ,125 11
2009 & 2011 108 ,707 12
2010 & 2012* 545 ,094 33
Three year period
2007 & 2010 618 ,269 7
2008 & 2011 969 ,055 6
2009 & 2012* 883 ,094 11
Four year period
2007 & 2011 1089 ,058 5
2008 & 2012* 1772 ,018 7
Five year period
2007 & 2012* 2145 ,043 5
* 2012 data were extrapolated
The above analysis identified significant mean differences (md) between the numbers of R&D hours performed by small Dutch SMEs in 2011 & 2012 (one year period, md [34] = 671, p < .013), 2008 & 2012 (four year period, md [7] = 1772, p < .018), and 2007 and 2012 (five year period, md [5] = 2145, p < .043). In all these compared periods a significant growth in the numbers of R&D hours took place.
We did not find evidence that the increase of the WBSO-grant resulted in more R&D activities of small Dutch SMEs in contrary to medium and large SMEs during the years of the increase
Furthermore we divided the sample in software development SMEs and physical development companies. This is a relevant distinction because the WBSO-grant is focused at labour cost and therefore the WBSO-grant supports labour intensive innovation stronger then capital intensive innovation, see section 2.1.
H3. The increase of the WBSO-grant resulted in more R&D activities of Dutch software development companies in contrary to physical development companies during the years of the increase.
Table 7: Number of R&D hours (software development SMEs) - Paired-Samples T-Test Compared years Mean difference (R&D hours) Sig. (2-tailed) N
One year period
2007 & 2008 -587 ,649 18
2008 & 2009 -548 ,758 22
2009 & 2010 1238 ,002 31
2010 & 2011 161 ,492 62
2011 & 2012* 292 ,260 54
Two year period
2007 & 2009 -1757 ,659 15
2008 & 2010 2522 ,031 21
2009 & 2011 1946 ,012 26
2010 & 2012* 670 ,098 53
Three year period
2007 & 2010 3553 ,069 14
2008 & 2011 3448 ,055 17
2009 & 2012* 2984 ,016 20
Four year period
2007 & 2011 4508 ,077 12
2008 & 2012 4925 ,066 13
Five year period
2007 & 2012* 6663 ,121 8
* 2012 data were extrapolated
The above analysis identified significant mean differences (md) between the numbers of R&D hours performed by Dutch software development companies in 2009 & 2010 (one year period, md [31] = 1238, p < .002), 2008 & 2010 (two year period, md [21] = 2522, p < .031), 2009 & 2011 (two year period, md [26] = 1946, p < .012), and 2009 & 2012 (three year period, md [20] = 2984, p < .016). In all these compared periods a significant growth in the numbers of R&D hours took place.
Table 8: Number of R&D hours (physical development SMEs) - Paired-Samples T-Test Compared years Mean difference (R&D hours) Sig. (2-tailed) N
One year period
2007 & 2008 -131 ,505 21
2008 & 2009 -720 ,395 25
2009 & 2010 266 ,286 22
2010 & 2011 185 ,558 20
2011 & 2012* 219 ,583 19
Two year period
2007 & 2009 -61 ,779 20
2008 & 2010 231 ,599 17
2009 & 2011 2069 ,299 19
2010 & 2012 159 ,671 16
Three year period
2007 & 2010 310 ,401 15
2008 & 2011 -201 ,563 14
2009 & 2012* 49 ,878 14
Four year period
2007 & 2011 -260 ,537 12
2008 & 2012* 382 ,469 10
Five year period
2007 & 2012* 366 ,507 9
* 2012 data were extrapolated
The above analysis did not identify significant differences between the numbers of R&D hours performed by Dutch physical development companies in periods from 2007 until 2012.
Comparison of the results of software development SMEs (Table 7 with relevant significant differences) and physical development SMEs (Table 8 with none significant differences) underline our expectations that software development SMEs are more stimulated by the increase of the WBSO-gran. As a result we accepted the H3. With this analysis we elaborated our previous accepted H1.
As described in our literature overview and method section, we handle the number of R&D employees as a proxy to measure the amount of innovation. Therefore we analysed the yearly numbers of R&D employees at Dutch SMEs with the next hypothesis. By this we tried to find more evidence of a growth in the amount of innovation during the increase of the WBSO-grant.
H4. The increase of the WBSO-grant resulted in additional R&D employees of Dutch SMEs during the years of the increase.
To analyse H4 we tested whether or not the number of R&D employees (proxy of operationalised measure of innovation) of Dutch SMEs increased during the years of the increase of the WBSO-grant. In Table 9 we present results of the Paired-Samples T-Test.
Table 9: Number of R&D employees - Paired-Samples T-Test Compared years Mean difference (number of R&D employees) Sig. (2-tailed) N
One year period
2007 & 2008 1,6 ,032 41
2008 & 2009 0,5 ,174 43
2009 & 2010 0,7 ,020 55
2010 & 2011 0,9 ,010 90
2011 & 2012 0,6 ,015 76
Two year period
2007 & 2009 2,4 ,016 34
2008 & 2010 1,2 ,113 40
2009 & 2011 2,0 ,005 46
2010 & 2012 1,7 ,001 73
Three year period
2007 & 2010 2,6 ,008 31
2008 & 2011 2,5 ,048 32
2009 & 2012 3,1 ,001 37
Four year period
2007 & 2011 3,7 ,028 25
2008 & 2012 4,0 ,015 25
Five year period
2007 & 2012 5,1 ,024 19
amount of innovation, no additional effect because of the increase of the WBSO-grant was recognized. Therefore we concluded that the increase of the WBSO-grant had no additional effect on the number of R&D employees at Dutch SMEs. As a result we rejected the H4.
We did not find evidence that the increase of the WBSO-grant resulted in additional R&D employees of Dutch SMEs during the years of the increase.
As presented in chapter 2 the increase of the WBSO-grant does not only aim at raising the amount of innovation in hours, the increase also intents to create additional investments by companies with respect to innovation. If indeed SMEs that receive the increased WBSO-grant initiate more innovative activities, this may indicate an effective policy. However it is equally possible that these innovative entrepreneurs obtain a payment reduction for activities that already were planned (see e.g. Brouwer & Van der Wiel - 2010), making public investments replace private investments (see substitution versus additionality in chapter 3). This potential effect is first analysed with the following hypothesis.
H5. The increase of the WBSO-grant resulted in additional R&D investments by Dutch SMEs during the years of the increase.
Table 10: Amounts of net R&D investments - Paired-Samples T-Test
Compared years Mean difference
(net R&D investments in Euro)
Sig. (2-tailed) N
One year period
2008 & 2009 -22935 ,336 42
2009 & 2010 6747 ,172 53
2010 & 2011 9882 ,128 87
Two year period
2008 & 2010 -14158 ,612 40
2009 & 2011 19926 ,057 43
Three year period
2008 & 2011 27850 ,201 31
The above analysis did not lead to significant outcomes. Therefore we did not find evidence that the increase of the WBSO-grant led to additional R&D investments by Dutch SMEs and we rejected the H5.
We did not find evidence that the increase of the WBSO-grant resulted in additional R&D investments by Dutch SMEs during the years of the increase.
So far our analyses with respect to H1, H2, H3, H4, and H5 present promising indications that the increase of the WBSO-grant may have led to a growth in the amount of R&D activities and led thus to more technological innovation. But at this point we also have to notice that although the numbers of R&D hours per SME significantly grew after the WBSO-grant increase, the R&D investments in Euro per SME did not. Nevertheless the amount of R&D investments also did not decrease. As shown in Figure 2 in chapter 2 normally the amount of R&D investments decrease during periods of economical recession. Therefore we concluded that we provide evidence of a positive effect of the increase of the WBSO-grant on the amount of R&D investments by Dutch SMEs.
Table 11: R&D investments - national figures Year R&D investments (in million Euro)
2007 5495 2008 5263 2009 4900 2010 5095 Source: CBS-statline 31-01-2012
These national figures include the R&D investments of companies that receive the WBSO-grant. The national figures underlined our conclusion about the positive effect of the increase of the WBSO-grant on the R&D investments of Dutch SMEs. Nevertheless we emphasise that this hypothesis requires further attention in a research with a larger sample.
The last analysis of section 4.2 is focused on the output side of innovation, the competitiveness / performance of Dutch SMEs. By this analysis we tried to find evidence of a potential effect of the increase of the WBSO-grant on this output side of innovation. Hereto we analysed the next hypothesis.
H6. The increase of the WBSO-grant resulted in better performance assessments for Dutch SMEs during the years of the increase.
As mentioned in section 3.1 we handled the productivity rate as the operationalised measure of a SMEs competitiveness. We tested whether or not the productivity of Dutch SMEs increased during the years of the increase of the WBSO-grant. In Table 12 we present results of the Paired-Samples T-Test.
Table 12: Productivity rate - Paired-Samples T-Test Compared years Mean difference
(productivity rate)
Sig. (2-tailed) N
One year period
2007 & 2008 1,7 ,251 38
2008 & 2009 -1,4 ,005 55
2009 & 2010 0,5 ,345 60
Two year period
2007 & 2009 0,4 ,700 33
2008 & 2010 -0,9 ,274 49
Three year period
