• No results found

University of Groningen Faculty of Economics and Business Master Thesis International Economics and Business

N/A
N/A
Protected

Academic year: 2021

Share "University of Groningen Faculty of Economics and Business Master Thesis International Economics and Business"

Copied!
28
0
0

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

Hele tekst

(1)

Faculty of Economics and Business

Master Thesis International Economics and Business

Name Student: Hui Chen Student ID number: 2106957

Student email: h.chen.4@student.rug.nl Date: Feb 19, 2014 (revised version)

Thesis supervisor: Sorin Krammer (DUI-520): m.s.s.krammer@rug.nl Thesis co-assessor: Dirk Bezemer ( DUI-513): d.j.bezemer@rug.nl

(2)

Determinants of Innovation Performance of Firms

Abstract

The objective of this paper is to investigate the determinants of innovation performance of firms. The data is covering over 10648 firms from all over the world in 19 industries. I use the negative binomial regression analysis to examine the relationships between innovation performance and the indicators at firm and country levels. The main findings are: R&D investment at firm and country levels has significant and positive effects on innovation performance of firms; FDI has a significant negative impact on innovation performance when only considering at country level. The influence of regulatory environment is negative. And larger firms are more likely to innovate. Finally, the costs of employees and secondary school enrollment may not be good measurements. Accordingly, some policy implications can be derived as well.

Key words: Innovation performance, R&D, FDI, Human capital, Regulatory

(3)

I. Introduction

Innovation is widely recognized to be a major driver of economic growth. A number of studies show that there is an endogenous relationship between innovation and economic growth. The results are controversial. But what we are sure is that innovation is very important. To clear what will affect the innovation is helpful to further study of the relationship between innovation and economic growth. Moreover, understanding how the innovation performance does a good job can assist the decision-makers of government and firm make strategy and policy better. The emphasis of this paper is the determinants of innovation performance at the firm level since actually firms do compete instead of nations.

There are only two sources of increasing the output of the economy: (1) increase more inputs that go into the productive process, or (2) innovation: turning new ideas into reality (OECD, 2004). In order to be competitive, firms must be of advantage of low cost or differentiation, both of which can be realized through innovation. Karasek (2012) states that innovation is perceived as a tool necessary to gain competitive advantage and one of the safest methods to defend one’s strategic position. And innovators can capture largest share of value (Dedrick et al, 2010), that is, the higher innovation performance firms possess, the more productive the firms will be. Firm specific characteristics can be sources of the different innovation performance, which I will discuss in the theory part.

However, the research of innovation performance of firms cannot be conducted

(4)

high-skilled intensive activities are more easily to get accessible to high skilled labors in this country. Productivity of this firm will be higher than the firms in other countries where human capital is relatively low. Therefore the contextual environment is also important for improving the innovation performance of firms. Some indicators of national context are also be presented in the next section.

The objective of this paper is to investigate the determinants of innovation performance of firms. Previous studies mainly focus on the determinants of innovation at either firm level or country level. The analysis combining multilevel approach is not much or not using the latest data. These two gaps will be filled in my paper. I examine the indicators at micro and macro levels at the same time by the latest available year 2012. The data is covering over 10648 firms from all over the world in 19 industries. I use negative binomial regression analysis to examine the relationships between innovation performance and the indicators at firm and country levels, which contains Research & Development (R&D) investment, Foreign Direct Investment (FDI) stocks, human capital, regulatory environment and firm size. 19 sector dummies are set to test whether industries will influence the innovation performance significantly. The main findings are: R&D investment at firm and country level has significant and positive effects on innovation performance of firms; FDI has a significant negative impact on innovation performance when only considering at country level. The influence of regulatory environment is negative. Larger firms are more likely to innovate. Finally, the costs of employees and secondary school enrollment may not be good measurements.

(5)

II Literature Review

According to reviewing a large number of previous literatures, I summarize six indicators that may impact the innovation performance of firms. The factors about firms’ characteristics are composed of R&D investment at the firm level, costs of employees, and firm size; while national context contains the R&D investment at the country level, FDI stocks, human capital, and regulatory environment. The following paragraphs will demonstrate and discuss them in details:

Absorptive capability and R&D investment

Improvement of innovation can be derived from the knowledge flows. Global business activities offer channels of knowledge flows. For example, MNEs need to create new (or improving) products to meet the requirement of foreign customers in the host market. At the same time, local firms can learn and absorb the advanced knowledge from foreign MNEs. However, knowledge transfer will not occur automatically. Liu and Buck (2007) point out only firms with a certain level of absorptive capacity are likely to take advantage of external technology spillovers. Absorptive capacity is defined as a firm’s ability to develop and improve its new products through the adaptation and application of the external technology stock (Cohen and Levinthal, 1990). Therefore a firm who has greater absorptive capability tends to innovate more.

(6)

Moreover, we can also determine absorptive capability from R&D investment at country level. The more a country invest in R&D, the higher level of technology will be in this country, the easier a firm in that country can acquire the knowledge than the firms in other countries with lower level technology. Unlike firms’ investment in R&D during the production, governments mainly concern investment in R&D infrastructures and researchers of institutions. Sufficient funds of R&D surely can provide a good R&D environment where innovative activities can be performed well. Therefore the first hypothesis is:

Hypothesis 1a: The innovation performance of firms is positively associated with

R&D investment at firm level.

Hypothesis 1b: The innovation performance of firms is positively associated with

R&D investment at country level.

FDI and Technology spillovers

(7)

Furthermore, inward FDI can stimulate local firms’ innovative activities through learning-by-doing or analyzing and observing the outputs of MNEs’ R&D projects (Liu & Buck, 2007), which is applicable to both horizontal and vertical FDI. Subsidiaries acquire the management skills and product designs from FDI. Other local firms can also learn the knowledge through interaction and cooperation with these subsidiaries, and finally improve their own innovation. Other researchers already find the evidences. For example, Branstetter (2006) asserts that inward FDI positively impact local firms’ innovation through knowledge spillovers. Foreign firms from advanced country bring up-to-date technology to host country. Local firms can also acquire the knowledge via imitation. More FDI stocks a country owns, more active the country participate in innovation. So my second hypothesis is presented as follows:

Hypothesis 2: The innovation performance of firms is positively associated with FDI

stocks of the country the firms locate in.

Human capital, Education level and Costs of employees

Human capital is the stock of competences, knowledge and cognitive skills that increase individual productivity in economic activities (Coleman, 1988). Many researchers have proved the positive relationship between human capital and economic growth in the past (Barro, 2001; Engelbrecht, 2003; Mankiw et. al., 1992). Human capital is also vital for firms. It is difficult to separate knowledge from those who possess it, and people involved in the organization are playing an increasingly important role in fostering innovation (Marvel & Lumpkin, 2007). Human capital can be regarded as an intangible asset of firms. Since innovation is a knowledge intensive activity, the human capital of employees will impact the competitiveness of the firm.

(8)

human resource to firms. Firms located in the country with higher human capital tend to own human resources more easily. Those firms located in the country with low human resource endowment will cost more to get human capital. Hence the third hypothesis is that:

Hypothesis 3: The innovation performance of firms is positively associated with the

human capital in the host country.

However, based on Schultz (1960)’s definition, Becker (1964) broaden the concept of human capital to including the on-the-job trainings. Actually, besides employees with high education, firms also prefer to hire trained and skilled workers because these employees possess high human capital as well. The previous studies on the human capital at the firm level are difficult in data collecting unless conducting specific individual survey. However, the wages can reflect employees’ ability. Workers with high capability can earn more. By the same token, firms are willing to pay more to talented and skilled workers. The wages paid by firms is included in the costs of employees in a firm. In addition, costs of employees also contain the training costs of employees. Vocational education and training (VET) is helpful to labors to gain practical working skills that can compensate labors with low education level. The more expenditure on training in a firm, the higher human capital the firm tends to own. Therefore the more spending on employees, the more firms are likely to innovate. The fourth hypothesis is that:

Hypothesis 4: There is a positive relationship between costs of employees and innovation performance of firms.

Regulatory environment

(9)

private actors in the economy1. Good regulatory environment guarantees market to operate effectively.

According to Bourreau and Doğan’s paper, regulation can impact innovative activities through two channels. First, regulations of price affect industry profits, hence the incentives to innovate. Second, price and entry regulation alter the terms of entry, therefore innovation decisions depending on new entry. So far literatures show the ambiguous results of regulatory effects on innovation. The effect of regulatory environment on innovation performance can be either positive or negative. On the one hand, regulations keep a certain level of openness of market to offer opportunity for innovation and also care the equity and fairness to protect the social welfare. On the other hand, regulations may set obstacles to technology diffusion. The following part will take the intellectual property rights as a specific example to illustrate how it influence the innovation performance of firms at positive and negative sides..

If intellectual property protection in one country is very strict, the innovators’ rights will be well protected. Firms recognize that their investments in innovation are guaranteed to be rewarded in the future, which stimulates firms reinvest in present market or develop new products. However, if local development is relatively lagging, the government is not willing to protect intellectual property right well because local firms need to improve their productivity via imitation. Therefore intellectual property protection can both motive and barrier to the innovation. Therefore the fifth hypothesis is that:

Hypothesis 5a: The innovation performance of firms is positively correlated to

regulatory environment.

Hypothesis 5b: The innovation performance of firms is negatively correlated to

regulatory environment.

1

(10)

Firm size

The relationship between firm size and innovation performance is mixed:

On the one hand, larger firms tend to innovate more since larger firms will have a greater pool of resources that help firms create non-duplicative knowledge (Gupta & Govindarajan, 2000). Hence large firms have the size advantages on internal knowledge, financial resources for innovation, sales base and market power (Cohen and Klepper, 1996).

On the other hand, small firms have little bureaucratic problems. They are flexible and they can react quickly to the market changes. Rapid decision-making let them adjust their products immediately to meet the customers’ requirements. Seizing the niche markets give them incentives to innovate. When a small firm develops to a larger one, it is relatively difficult to communicate within the firm. And long decision-making and slow reaction time may lead firm missing some sound business chances. The firm is less likely to innovate when it “grows up”.

These two opposite assumptions lead to two different hypotheses below:

Hypothesis 6a: Firm size has a positive relationship with innovation performance of

firms.

Hypothesis 6a: Firm size has a negative relationship with innovation performance of

firms.

III. Data and Method

(11)

Dependent variable

Innovation Performance is my dependent variable, which represents how well a firm

has done on its innovation. It is measured by the number of patents of each firm. As Filatotchev et al (2011) suggested, patents are directly related to innovative outputs. The data is from Orbis2, an organization engaged in collecting company information across the globe. The number of patents varies from 1 to 18647.

Independent Variables

R&D expenditure, representing absorptive capability of the firm. The value of R&D

expenditure is measured in percentage of firm operating revenue at firm level from orbis and the percentage of GDP at country level from World Databank3. A two-year lag is expected because R&D expenses may not innovate immediately to help improve innovation performance4.

Costs of employees are the costs of employees of each firm, divided by firms’

operating revenue in order to correcting for the firm size. The data is also from the Orbis.

Data of FDI stocks is the inward foreign direct investment stocks, measured in percentage of GDP, from UNCTADstat (2013)5 with a four year lag. FDI stocks are presented at book value, reflecting prices at the time when the investment was made6.

The data of Human Capital is the gross secondary school enrollment ratio. This ratio is the total enrollment in secondary education, regardless of age, presented as a 2 https://orbis.bvdinfo.com.proxy-ub.rug.nl/version-2013115/Home.serv?product=orbisneo&loginfromcontext=ipad dress 3 http://databank.worldbank.org/data/views/variableselection/selectvariables.aspx?source=world-development-indic ators 4

Scholars have different views on how long the time lag should be (see Appendix 1). The time lag may different from country to country. Two years lag seems to be preferable. So the data of R&D expenditure is selected form 2010’s data.

5

http://unctadstat.unctad.org/ReportFolders/reportFolders.aspx?sRF_ActivePath=P,5,27&sRF_Expanded=,P,5,27 6

(12)

percentage of the population of official secondary education age. The value maybe larger than 100% since it also includes over-aged and under-aged students because of early or late school entrance and grade repetition7. A 4-year lag is applied to human capital since people in secondary school will generally be entering the labor market some time later and will not be productive for around 4 years. Again, data is cited from World Databank.

Finally, Regulatory Quality is used to measure countries’ regulatory environment. I also accept a two year lag. It is an index that comes from Worldwide Governance Indicators (WGI)8 in 2010. Regulatory Quality reflects perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development. The estimate of Regulatory Quality ranges from approximately -2.5 (weak) to 2.5 (strong) governance performance. Namely, closer the value to 2.5 is, higher the quality of regulation of country where the firms are located will be.

Control variables

Firm Size is my control variable. It is presented as the number of employees of each

firm, measured by natural logarithm. The data is also from Orbis.

Table 1 is the descriptive statistics of each variable. And appendix 2 is the correlations between all variables, which shows their correlations are very low. Therefore any of the independent variables will not be taken reliably as a proxy for any of the others. No one needs to be deleted.

7

(13)

Table 1

Descriptive Statistics

Variable Obs Mean Std.dev. Min Max

Innovation Performance 10648 246.755 1013.018 1.000 18647

R&D expenditure (% of total revenue) 3796 3.296 5.899 0.000 79.660

R&D expenditure (% of GDP) 10089 2.500 0.779 0.194 4.347

Human Capital 10053 97.765 9.999 61.931 126.742

FDI stocks (% of GDP) 10645 22.011 24.136 4.194 226.397

Regulatory Quality 10645 1.141 0.575 -1.610 1.890

Costs of employees (% of total revenue) 4535 16.039 12.263 0 97.580

Log of number of employees 10648 3.306 0.695 0 6.342

Method

I consider negative binomial regression analysis for cross-sectional data to examine the determinants of innovation performance. Negative binomial regression is for modeling count variables, usually for over-dispersed count outcome variables. OLS regression is not helpful in this case. OLS regression may lose data due to undefined values generated by taking the log of zero. And it may also be lack of capacity to model the dispersion9. However, negative binomial regression can be used for over-dispersed count data when the conditional variance goes beyond the conditional mean. My dependent variable Innovation Performance is count data and its histogram shows its distribution is strongly skewed to the right and dispersed, which means negative binomial regression should be accepted. In a word, the advantage of negative binomial regression overweighs the merit of OLS regression in my case.

Moreover, I consider that the economic level of host country may impact the innovation performance of firms in those countries. I also divide companies into two groups to regress the model respectively, according to host countries’ economic level. Namely, one group contains the firms from developed countries and the other group consists of firms from developing countries. Appendix 3 shows the company breakdown by country.

9

(14)

In addition, the 10648 firms are recognized to 19 industries (see Appendix 3) which are codified to 19 dummy variables, in order to examine whether different sectors will impact the results significantly.

IV. Results

Overall Impact

Table 2 presents the estimated coefficients of the independent variables and their corresponding significance levels. All the firms’ data are examined. At first, I run the regression only with firm level variables. Second, only country level data are tested. Third, I incorporate firm and country level variables into one regression. The next three regressions are re-examined like the former three by adding the sector dummy variables.

As seen from the table, in all specifications, the estimates of R&D expenditure at the firm level and R&D expenditure at the country level are both significantly positively associated with the innovation performance at 1% level. Hypothesis 1a and 1b are accepted. FDI stocks has a negative relationship with innovation performance when only considering country level indicators. The signal is opposite to my expectation. So

Hypothesis 2 is rejected. The coefficients of Human capital are not significantly

associated with innovation performance under any situations. Therefore Hypothesis 3 is not supported. The coefficients of Costs of employees are all around -0.025, which means costs of employees have negative effects on innovation performance. So

Hypothesis 4 is not accepted. The coefficient of Regulatory Quality has a negative and

(15)

positively significantly. Therefore Hypothesis 6a is supported while Hypothesis 6b is rejected. Moreover, the sector dummy uncovers that locating in sector 1, 2, 6, 7, 8, 9, 11, 12, 13, 14, 15, and 16 will significantly influence firm’s innovative performance when only examining at the country level indicators.

(16)

Sector 15 (omitted) -1.259*** (omitted)

Sector 16 (omitted) -3.066*** (omitted)

Sector 17 0.025 0.007 -0.313

Sector 18 -1.919 -0.413 -2.001

Sector 19 (omitted) (omitted) (omitted)

Chi statistic 622.25*** 1638.00*** 478.10*** 788.81*** 3164.25*** 626.91***

Valid N 1461 9880 1074 1460 9863 1073

Dependent variable: Innovation Performance

***, ** and * indicate significance are at 1%, 5% and 10% level

Impact in Developed country group

Table 3 is the regression results with firms from developed countries.

According to the table, the results of R&D expenditure at the firm level and R&D expenditure at the country level are similar to the overall impact. Hypothesis 1a and

1b are accepted. FDI stocks also has a negative relationship with innovation

performance when only considering country level indicators. So Hypothesis 2 is still rejected. The coefficients of Human capital are not significantly associated with innovation performance under any situations, which is the same as overall impact. Therefore Hypothesis 3 is not supported. The coefficients of Costs of employees are all around -0.03, which means costs of employees have negative effects on innovation performance. So Hypothesis 4 is not accepted. The coefficient of Regulatory Quality has a negative and significant effect on innovation performance of firms except for column 12. So Hypothesis 5a is rejected while 5b is partially accepted. Firm size also shows it affects firms’ innovation performance positively significantly. Therefore

Hypothesis 6a is supported while Hypothesis 6b is rejected. Furthermore, Sector 2, 6,

7, 8, 9, 11, 12, 13, 14, 15, and 16 turn out to have significant effect on the results when only adding country level indicators.

Table 3

Developed countries

Variables Coefficients

(17)

R&D expenditure (% of total revenue) 0.073*** 0.072*** 0.054*** 0.055*** Costs of employees (% of total revenue) -0.033*** -0.028*** -0.033*** -0.026*** Log of number of employees 1.206*** 1.143*** 1.177*** 1.102*** Regulatory Quality -0.832*** -0.499** -0.708*** -0.316 R&D expenditure (% of GDP) 1.109*** 0.549*** 1.033*** 0.506*** Human Capital -0.001 0.006 -0.001 0.014 FDI stocks (% of GDP) -0.002** 0.000 -0.003*** -0.002 Sector 1 -1.176 0.119 -1.180 Sector 2 -2.064* -0.537*** -2.110* Sector 3 -2.003* -0.110 -2.100* Sector 4 -1.190 0.126 -1.641 Sector 5 -0.482 0.251 -0.811 Sector 6 0.011 1.200*** -0.282 Sector 7 -0.209 0.598*** -0.630 Sector 8 0.194 1.219*** -0.162 Sector 9 -1.840 -0.861*** -2.257** Sector 10 -1.043 -0.080 -1.581 Sector 11 -0.771 -0.478*** -1.009 Sector 12 -2.535* -2.079*** -3.064** Sector 13 -1.249 -1.330*** -1.570 Sector 14 -0.604 1.182*** -0.869

Sector 15 (omitted) -1.315*** (omitted)

Sector 16 (omitted) -2.550*** (omitted)

Sector 17 -0.328 0.064 -0.854

Sector 18 -1.802 -0.222 -1.895

Sector 19 (omitted) (omitted) (omitted)

Chi statistic 326.73*** 1324.70*** 360.20*** 448.08*** 2712.63*** 476.48***

Valid N 825 8215 784 825 8201 784

Dependent variable: Innovation Performance

***, ** and * indicate significance are at 1%, 5% and 10% level

Impact in Developing country group

(18)

Based on the table, the signal and significance of coefficients have two differences, compared to table 1 and 2. R&D / GDP and regulatory environment show no association with innovation performance of firms. All the sectors can significantly affect firm’s innovative performance when only examining at the country level indicators. One point needs to be mentioned. The regression 18 fails because the valid sample is only 288. I have 26 independent variables and as a rule of thumb the sample size should be more than 390. 228 are far less than the standard. Hence, regression 18 is not under consideration.

(19)

Sector 12 -3.947** -7.098*** -- Sector 13 -2.001** -4.962*** -- Sector 14 -0.373 -2.751** -- Sector 15 (omitted) -4.272*** -- Sector 16 (omitted) -7.005*** -- Sector 17 (omitted) -3.898*** --

Sector 18 (omitted) (omitted) --

Sector 19 (omitted) (omitted) --

Chi statistic 151.32*** 150.89*** 162.25*** 205.27*** 480.09*** --

Valid N 323 1598 289 322 1595 288

Dependent variable: Innovation Performance

***, ** and * indicate significance are at 1%, 5% and 10% level

Summary

In order to observe the results clearly, I summarize them in Table 5.

Table 5

Hypotheses Independent variables Expected signal Overall Developed Developing

H1a R&D/revenue + Y Y Y

H1b R&D/GDP + Y Y N

H2 FDI stocks + N (p.neg.sig) N (p.neg.sig) N (p.neg.sig)

H3 Human capital + N N N

H4 Costs of employees + N (neg.sig.) N (neg.sig.) N (neg.sig.)

H5a Regulatory environment + N N N H5b - Y (p.sig) Y (p.sig) N H6a Firm size + Y Y Y H6b - N N N

NOTE: Y: support & significant; N: reject; P: partially; neg: negative; sig: significant

V. Discussion and Conclusion

(20)

support to research institutes and their scientists. Scientific innovation improvement is a comparative advantage for firms located in that country. Advanced knowledge is easier to get accesses. Low transaction cost also brings firms a cost advantage.

Unfortunately, FDI stocks do not show robust results on innovation performance of firms. Considering only country level indicators, it shows a negative relationship. It is still opposite to expectation. Chen (2007) also gains a negative relationship and gives a possible reason that may explain this issue. She thinks MNEs is more likely to possess up-to-date technology and rich management skills. MNEs may gain monopoly power in local market. High competition makes firms be willing to purchase the technologies directly from abroad instead of spending large money on their own R&D. Therefore local firms’ innovative activities decrease. Moreover, von Hipple (1994) states that information has stickiness characteristic. But it does not mean FDI only has negative impacts. Actually FDI can be classified into Greenfield FDI and Merge and Acquisition (M&A). Bertschek (1995), as well as Blind and Jungmittag (2004), claim that Greenfield investments may induce positive effects via increased competition as the results of previous theory section indicates. Therefore regulations that encourage Greenfield FDI should be favored.

The results also show that human capital has no effect on innovation performance. It is totally contrary to the theory and other literature. I think the issue should be the measurement of the human capital. Secondary school enrollment fails to measure human capital. Secondary school attainment is better because it guarantee knowledge is really successfully acquired by people. But data of attainment in World Bank is lacking. And actually educational level is not the only one that can represent one country’s human resource. Working experience play a role at the same time.

(21)

possess high capability. Furthermore, firms in different sectors may need different type of labors. For example, firms in pharmaceutical industry will be engaged in more knowledge intensive activities; while assembly firms are engaged in more labor intensive activity. Assembly firms cannot get the same returns as pharmaceutical firms if assembly firms give the same high payments to the employees like pharmaceutical firms. And higher costs of employees may account for overmuch profit, which leads to decrease in R&D spending. Hence, the potential solution is to gain the detailed personnel information of firms and test the model again.

The impact of regulatory environment on innovation performance is negative, which explains that establishing a high quality of regulation is harmful to innovation performance of firms. Strict protecting regulation may restrict the technology spillovers. Imitation becomes more difficult for lagging firms. The policy-makers should make the suitable regulation depending on different actual situation. Sound regulations should be beneficial to technology diffusion instead of being barriers.

At last, firm size has a positive effect on the innovation performance of firms. Larger firms have richer resource and higher competence, which contributes to innovative activities. So smaller firms perform less innovative than larger ones.

In this paper, I also divide sample into two groups, firms from developed countries and firms from developing countries. The results of developed countries group are similar to the results of the whole sample but results of developing countries have little differences on macro R&D investment and regulatory environment. According to the breakdowns by countries, number of firms from developing countries is much less than the number of firms from developed countries. And Chinese firms amount to the major share of firms. It may cause a selection bias.

(22)

innovation performance at firm, sector and country level simultaneously and the selection of data may be somewhat out of date. However, my data is covering over 10648 firms from all over the world in 19 industries, selecting from the latest years. In the methodology part, I use the negative binomial regression analysis to examine the relationships between innovation performance and the indicators. The factors of innovation performance of firms are finally determined to be R&D investment at firm and country level, firm size and regulatory environment.

The majority of the results are consistent with my expectations. The main findings are: R&D investment at firm and country level has significant and positive effects on innovation performance of firms; FDI has a significant negative impact on innovation performance when only considering at country level. The influence of regulatory environment is negative. Larger firms are more likely to innovate. Finally, the costs of employees and secondary school enrollment may not be good measurements. Accordingly, policy implications can be derived as follows:

At country level, first, government support sufficient funds to educational institutes and scientists. Secondly, Greenfield FDI ought to be encouraged. Thirdly, regulations should be made properly and should not be barriers to technology diffusion. At the firm level, first, managers can increase the ratio of investment in R&D amounting to the total revenues. Secondly, firms in different country and different sectors may need different level of innovation. They may need different determinants.

(23)

measurement for innovation performance since not all the innovations are patented. A large number of intangible innovations are still not counted, such as management skills. Therefore, a more comprehensive indicator is required to fully capture firm’s innovative performance, which might be the subject of future research.

Acknowledgement

I would like to express my deepest gratitude to my supervisor Mr. Krammer who gave me valuable instructions and insightful comments on my thesis. I am also grateful to Mr. Harchaoui for the advice of methodology section and I do appreciate Mr. Bezemer for all his suggestions. Finally, I thank all the instructors who guided me during the past one year for their teaching and helping.

Reference

Barro, Robert J. (2001). Human capital and growth. American Economic Review, Vol: 91, No. 2, pp. 12-17.

Becker, G. S. (1964). Human capital. New York: Columbia University Press.

Bertschek, I. (1995), Product and Process Innovation as a Response to Increasing Imports and Foreign Direct Investment, Journal of Industrial Economics 43(4): 341–357.

Blind, K. and Jungmittag, A. (2004), Foreign Direct Investment, Imports and Innovations in the Service Industry, Review of Industrial Organization 25 (2): 205-227.

Bourreau, M., and Doğan, P., Regulation and Innovation in the Telecommunications

Industry, forthcoming Telecommunications Policy. Retrieved from

(24)

Branstetter, L., (2006). Is foreign direct investment a channel of knowledge spillovers? Evidence from Japan’s FDI in the United States. Journal of

International Economics 68, 325–344.

Chen, Yufen, 2007. Impact of foreign direct investment on regional innovation capability: a case of China, Journal of Data Science 5(2007), 577-596.

Cohen, W.M. and D.A. Levinthal, (1989), Innovation and Learning: The two faces of R&D, Economic Journal, 99 (397), 569-596.

Cohen, W., Levinthal, A., (1990). Absorptive capability: the new perspective learning and innovation. Administrative Science Quarterly, 35: 128–152.

Cohen, W. and Klepper, S., (1996), A Reprise of Size and R&D, Economic Journal, 106, 925-951.

Coleman, J. S., .Social capital in the creation of human capital. American Journal of

Sociology, nr.94, 1988, p. S95-S120

Dedrick, J., K.L.Kraemer and G. Linden (2010), “Who Profits From Innovation in Global Value Chains? A Study of the iPod And Notebook PCs”, Industrial and

Corporate Change, 19 (1): 81-116.

Engelbrecht, H. J. (2003). Human Capital and Economic Growth: Cross-Section Evidence for OECD Countries, Economic Record, No.79 (special issue), pp 74–84.

Englander, S., Evenson, R. and H. Masaharu (1988). R&D, Innovation and the Total Factor Productivity Slowdown, OECD Economic Studies 11:7-42

Filatotcheva, I., Liu, X., Lu, J., and Wright, M. (2011). Knowledge spillovers through human mobility across national borders: Evidence from Zhongguancun Science Park in China. Research Policy: 40 (2011) 453-462.

Gorg, H., and Strobl, E., (2005). Spillovers from foreign firms through worker mobility: an empirical investigation. Scandinavian Journal of Economics 107(4): 693-709

Greif S. (1985), Relationship between R&D expenditure and patent applications,

World Patent Information, Vol. 7, N°. 3, pp. 190-195, 1985

(25)

Gupta, A. K., and Govindarajan, V. (2000). Knowledge flows within multinational corporations, Strategic Management Journal, 21 (4): 473-496.

Hall BH, Griliches Z, Hausman JA (1986) Patents and R&D: Is there a lag?

International Economic Review 27: 265-283.

Karasek, A. (2012). INNOVATION AND EXPORT PERFORMANCE: EVIDENCE FROM LUBLIN VOIVODESHIP. Management, Knowledge, and Learning, International Conference 2012. Retrieved from

http://www.issbs.si/press/ISBN/978-961-6813-10-5/papers/ML12_200.pdf

Kondo M., R&D (1999), Dynamics of creating patents in the Japanese industry,

Research Policy 28, 1999: 587-600

Krugman, P. (1994). Competitiveness: A dangerous obsession. Foreign Affairs, 73(2), 28.

Liu, X. and Buck, T. (2007). Innovation performance and channels for international technology spillovers: Evidence form Chinese high-tech industries. Research

Policy, 36(3): 355-366.

Liu, X., Lu, J., Filatotchev, I., Buck, T., and Wright, M. (2010). Returnee entrepreneurs, knowledge spillovers and innovation in high-tech firms in emerging economies. Journal of International Business Studies, 41 (7): 1183-1197

Mankiw, N. G., Romer, D. and Weil, D. N. (1992). A Contribution to the Empirics of Economic Growth, Quarterly Journal of Economics, pp. 407-437.

Marvel, M., & Lumpkin, T. (2007). Technology entrepreneurs’ human capital and its effects on innovation radicalness. Entrepreneurship Theory and Practice, 31(6): 807–828.

OECD, (2004). Retrieved from http://www.oecd.org/cfe/tourism/34267902.pdf

Park, W. (1995), International R&D Spillovers and OECD Economic Growth,

Economic Inquiry 33: 571-591.

Schultz, T.W., (1960). Capital Formation by Education, Journal of Political

Economy, Vol. 68, No. 6 (Dec., 1960), pp. 571-583

(26)

Appendix

1. R&D expenditure time lag

Time Lag Author(s) Observation

no lag Hall et al., (1986) US

between 1 and 2 years Greif, (1985) Germany

between 1.5 and 1.7 years

Kondo, (1999) Japan

3 years

Englander et al. (1988); Park (1995) 2. Correlation of Variables Innovation Performance R&D expenditure (% of total revenue) R&D expenditure (% of GDP) Human Capital FDI stocks (% of GDP) Regulatory Quality Costs of employees (% of total revenue) Log of number of employees Innovation Performance 1 R&D expenditure (% of total revenue) 0.094 1 R&D expenditure (% of GDP) 0.0273 -0.1024 1 Human Capital 0.0281 0.0493 -0.0743 1 FDI stocks (% of GDP) 0.0149 0.1394 -0.3821 0.4185 1 Regulatory Quality 0.0251 0.1538 -0.1855 0.6102 0.4503 1 Costs of employees (% of total revenue) 0.0447 0.1817 -0.3338 0.3403 0.3114 0.4276 1 Log of number of employees 0.2973 -0.0085 -0.3215 0.1404 0.199 0.0655 0.3644 1

3. Company breakdown by country and sector

Code Sector Number of

Companies

01 Primary sector 230

02 Food, beverages, tobacco 534

(27)

04 Wood, cork, paper 273

05 Publishing, printing 177

06 Chemicals, rubber, plastics, non-metallic products 1601

07 Metals & metal products 788

08 Machinery, equipment, furniture, recycling 2999

09 Gas, Water, Electricity 313

10 Construction 411

11 Wholesale & retail trade 1096

12 Hotels & restaurants 60

13 Transport 202

14 Post & telecommunications 157

15 Banks 176

16 Insurance companies 14

17 Other services 1236

18 Public administration & defense 11

19 Education, Health 165

Total 10648

Number of companies Total

(28)

Referenties

GERELATEERDE DOCUMENTEN

As theoretically predicted, the empirical results show that firm performance is negatively related to product diversification, but positively related to

Majocchi and Zucchella (2003) do not find any evidence for a relation between exporting and performance, which can be attributed to a sample made up of only exporters according

In another article (Ito & Lee, 2005b) the authors compared the impact of 9/11 and its after-effects on airline demand in Australia, Canada, Europe, Japan and the U.S. Using the

Official election data has been extracted both from the historical archive of the Ministry for Internal Affairs (Ministero degli Affari Interni, s.d.) and the Global Election

When a set of control variables are added (2), the significance for middle- income share becomes stronger (0.1%) and when control variables are added for industrial jobs (4),

The link between regional competitiveness and the development of human capital is primarily a result of resources gained because of the region’s competitive position vis-à-vis

The average deposit rate of the other banks operating in each location is calculated which then is used to create an instrument which exogenously influences the deposit rate for

Results in table 6 (model 5) imply that there is no difference between state owned firms and private owned firms on basis of export propensity when there are control variables