Venture capital, Ownership concentration and Enterprise R&D investment

Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science 91 (2016 ) 519 525 Information Technology and Quantitative Management (ITQM 2016) Venture capital, Ownership concentration and Enterprise R&D investment Hongxing Wen a, Kui Xia a, * a School of Business, Sichuan University, #24 South of the ring road, Chengdu 610064, P.R. China Abstract We empirically examine the relationship among venture capital (VC), ownership concentration and enterprise R&D investment. Drawing on the data of China s GEM from 2010 to 2014, we find that VC and enterprise R&D investment are significantly positively correlated while ownership concentration and enterprise R&D investment are significantly negatively correlated. Further research shows that in the VC-backed companies ownership concentration has no significant impact on R&D investment. 2016 The The Authors. Published Published by Elsevier by Elsevier B.V. This B.V. is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Selection and/or peer-review under responsibility of the organizers of ITQM 2016 Peer-review under responsibility of the Organizing Committee of ITQM 2016 Keywords: VC; Ownership concentration; Enterprise R&D investment 1. Introduction GEM (Growth Enterprises Market) has been playing an important role in China's capital market since it was officially launched in October 30, 2009. Companies listed on the GEM are mostly engaged in high-tech business with high growth potential. For these enterprises, technological innovation capability is the core competitiveness and is essential to the growth and development of the enterprises [1]. R&D (research and development) investment is the key factor to improve the technological innovation capability, so R&D investment decision is one of the most important decisions for high-tech enterprises. Most of the GEM listed companies are small and medium-sized enterprises (SMEs). Their performance is not outstanding, and it is difficult for them to get the funds by traditional financing methods such as bank loans to support their R&D investment. As a new way of financing, venture capital (hereafter VC) can provide R&D investment funds for GEM companies, and the professional management experience of venture capitalists is helpful to improve the value of the company through technological innovation [2]. The largest shareholder generally has a great control over the company and is able to affect the company's R&D decisions through the form of the general meeting of shareholders. Can VC effectively promote the R&D investment of GEM listed * Corresponding author. Tel.: +8615828062455. E-mail address: 1162835997@qq.com. 1877-0509 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the Organizing Committee of ITQM 2016 doi:10.1016/j.procs.2016.07.133

520 Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 companies? What is the impact of ownership concentration on enterprise R&D investment? How the interaction between VC and ownership concentration affect R&D investment of GEM listed companies? These are the issues to be studied in this paper. 2. Literature Review And Hypotheses 2.1. The influence of VC on enterprise R&D investment VC can not only provide financial support for innovative SMEs, but also bring internal and external oriented value-added services, such as the company's standardization, business expansion, development strategy, supervision, consultation and so on [3]. VC pays more attention to the construction and maintenance of longterm competitiveness of enterprises, and focuses on investment and quality of innovative activities [4], which is conducive to the development of entrepreneurial firms to improve R&D investment [5-6]. Using data from 20 US industries from 1983 to 1992, Kortum and Lerner [7] demonstrated the relationship between VC and innovation. The results showed that: in terms of the proportion of patents, VC-backed enterprises were higher than the non-vc-backed ones. Chesbrough [8] found that the existence of VC and enterprise R&D investment had a very significantly positive correlation. Gou and Dong [9] empirically studied the listed companies of China s SME board and found that in terms of technological innovation, the performance of VC-backed enterprises was significantly better than the enterprises with no VC participation whether from the perspective of R&D investment or from the perspective of the number of patents. Xie [10] drew a similar conclusion based on the sample data of 189 IPO companies listed on the China s GEM. Accordingly, the following hypothesis is presented: H1: VC is significantly positively related to enterprise R&D investment. 2.2. The impact of ownership concentration on enterprise R&D investment Due to the difference of research method, perspective and data, research conclusions of the existing literature differ greatly when it comes to the relationship between ownership concentration and R&D investment. The main relationships between the two include positive correlation [1], negative correlation [11], U type relationship [12] and inverted U type relationship [13]. However, the technological innovation projects of the enterprises are often risky and innovative investment has the characteristic of irreversible. From the perspective of risk aversion, the more concentrated ownership, the greater special risk the major shareholders of the company will bear, the less they are willing to invest in technology innovation projects [14]. In addition, with the increase in the proportion of the shares of large shareholders, when making decisions, they will have more decision-making power. Thus, large shareholders are likely to take advantage of their agent identity for minority shareholders to infringe the interests of small shareholders. The infringement will lead to the loss of efficiency of the company's decision-making which has an adverse effect on R&D investment [15]. Suk Bong Choi et al. [16] found when a country was in the economic transition, the high degree of ownership concentration was not necessarily a positive effect on the innovation behavior through empirical study. Ortega- Argilés [17] pointed out that high ownership concentration hindered companies to improve R&D investment and was not conducive to the development of R&D output. Therefore, we put forward the following hypothesis: H2: There is a significant negative correlation between ownership concentration and enterprise R&D investment.

Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 521 2.3. The impact of the interaction between VC and ownership concentration on enterprise R&D investment VC emphasizes the high growth of entrepreneurial enterprises and pays attention to the innovation of the enterprises. VC is willing to supervise the enterprises to carry out R&D investment. Xiong and Chen [18] pointed out that high-tech enterprises had special corporate governance structure due to VC accounted for a considerable proportion of equity. The special governance mechanism could better deal with conflicts and interests of all stakeholders so that high-tech enterprises could get the driving force for sustainable development. Thus, it can be predicted that compared to non-vc-backed companies, in the VC-backed ones ownership concentration has a smaller negative impact on enterprise R&D investment. So we put forward the following hypothesis: H3: The participation of VC can significantly reduce the negative impact of ownership concentration on enterprise R&D investment. 3. Research method 3.1. Sample and Data Collection This paper selected 2010-2014 China s GEM listed companies as the initial sample, and selected samples in accordance with the following standards: (1) excluded the companies of financial industry; (2) removed the samples of ST or data with missing key values. Ultimately we got 458 observation values, and the observed values for each year were shown in table 1. Table1. The number of observations in each year Year 2010 2011 2012 2013 2014 Observed number 51 79 94 107 127 For the definition of VC, we referred to Wu [19], Fu [20], Wang [21], and define it according to the following steps: (1) if the names of the company's shareholders contained venture capital, then the shareholders were taken as VCs; (2) if the names of the company's shareholders did not contain venture capital, then we referred to the 2012 edition of the "China venture capital and private equity investment institutions" compiled by the Zero2IPO database and the directory of VC included in "Chinese venture investment development report 2013", if any of the shareholders of the company entered any one of these two lists, then it was considered as VC; (3) if the shareholders names were not in the above lists, then we consulted the company's shareholders and if the main business of the shareholder of the company was venture capital business, then the shareholder was recognized as VC. If any of the company's shareholders was VC, then we took the company as a VCbacked company. All the financial data were collected from the CSMAR database. 3.2. Variable definition 3.2.1 Dependent variable Draw lessons from Yang and Yuan [22], we use the ratio of natural logarithm of R&D expenditure to natural logarithm of the main business income to measure the enterprise's R&D investment. 3.2.2 Independent variable In order to measure whether the enterprise is VC-backed enterprise or not, this paper introduces VC as the

522 Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 explanatory variable; In order to explain the impact of ownership concentration on enterprise R&D investment, we set up the explanatory variable of ownership concentration and use the proportion of the first largest shareholder to measure it. Table2. Variable definition Variables symbol Variable definition R&D investment VC participation Ownership Concentration R&D VC TOP1 The ratio of natural logarithm of R&D expenditure to natural logarithm of the main business income If VC participates in a company, then the value of VC is 1, otherwise the value is 0 The proportion of the first largest shareholder Company size SIZE Natural logarithm of total assets Asset liability ratio LEV The Ratio of total assets to total liabilities CEO duality DUAL When the chairman is also the CEO, the value of DUAL is 1, otherwise the value is 0 Industry IND According to the 2012 industry classification made by the China s Securities Regulatory Commission and combined with the actual situation, we set 10 industry dummy variables Year YEAR We set 4 annual virtual variables 3.2.3 Control variables According to previous literature, we select the company size, asset liability ratio, whether CEO separates from chairman (CEO duality), the year and the industry as control variables. The definition of the above variables is shown in table 2. 3.3. Regression model In order to test the hypothesis 1, we construct the following regression model: R&D = β 0 + β 1 VC +β 2 SIZE +β 3 LEV + β 4 DUAL + IND + YEAR + ε (1) If the coefficient of β 1 is significantly positive, then the VC participation is conducive to the development of enterprise to increase R&D investment. In order to test hypothesis 2, we establish the following model: R&D = β 0 + β 1 TOP1 +β 2 SIZE +β 3 LEV + β 4 DUAL + IND + YEAR + ε (2) If the coefficient of β 1 is significantly negative, then it shows that ownership concentration is negatively related to the R&D investment of enterprises. That is, the increase of ownership concentration is not conducive to the enterprise R&D investment. In order to test the hypothesis 3, the following model is established: R&D = β 0 + β 1 VC +β 2 TOP1 +β 3 VC TOP1 + β 4 SIZE + β 5 LEV + β 6 DUAL + IND + YEAR + ε (3)

Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 523 We judge whether VC can able to regulate and how to regulate the impact of ownership concentration on enterprise R&D investment by the coefficient of VC TOP1. 4. Empirical analysis result 4.1 Descriptive statistical analysis Table3. Descriptive statistics of the main variables Variables Min Max Mean Std. dev. R&D 0.5542 1.0119 0.7997 0.0743 VC 0.0000 1.0000 0.4100 0.4920 TOP1 0.0496 0.6150 0.3030 0.1204 SIZE 19.3913 23.1543 20.9356 0.6433 LEV 0.0110 0.6875 0.2059 0.1492 DUAL 0.0000 1.0000 0.4300 0.4960 Table 3 lists the descriptive statistics for the main variables. As we can see from the table 3, the standard deviation of R&D is 0.0743, which shows that R&D investment of the GEM listed companies is relatively uniform; The mean value of VC is 0.41, which indicates that VC has participated in about 41% of the GEM listed companies; The mean value of TOP1 is 0.303, which indicates that the ownership concentration of the GEM listed companies is high, and the standard deviation of it is 0.1204, indicating that there is a small difference between companies; The mean value of DUAL is 0.43, which means that about 43% of the GEM listed companies chairmen are also CEOs. 4.2 Regression analysis Table 4 summarizes the regression results of the above 3 models. From the regression results of the model (1), the coefficient of VC is 0.0205 with a significant level of 1%. This shows that companies involved in VC pay more attention to R&D investment, hypothesis 1 is validated. The coefficient of TOP1 in the model (2) is - 0.0581 at the 5% level of significance which shows that ownership concentration and enterprise R&D investment are significantly negatively correlated. That is, with the increase of ownership concentration, companies will reduce R&D investment, which is consistent with the hypothesis 2. Model (3) introduces the cross terms of VC and TOP1 to explore the impact of the interaction between VC and TOP1 on R&D investment. The empirical result shows that the coefficient of VC TOP1 is -0.0167 but not significant. This shows that ownership concentration has no significant impact on R&D investment, which is not consistent with the hypothesis 3. The possible reason is that the ownership concentration of the GEM listed companies is generally high, coupled with the largest shareholder in part of the companies is VC, which leads to the limited moderating effect of VC on the impact of ownership concentration on R&D investment. Moreover, in terms of control variables, the regression coefficients of DUAL, LEV and SIZE of the three models are significant at 1% level. In the above three models, the coefficients of SIZE are all positive which shows that the larger the scale of the company is, the more R&D investment will be. The coefficients of LEV are all negative, which indicates that with the increase of the level of debt, companies tend to reduce R&D investment. The coefficients of DUAL are all positive. This shows that in China s GEM listed companies CEO duality is conductive to the R&D investment of the enterprises.

524 Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 Table4. Summary of empirical results 5. Conclusions Variables Model(1) Model(2) Model(3) VC 0.0205 *** 0.0232 (2.9964) (1.2087) TOP1-0.0581 ** -0.0366 (-2.1105) (-1.0269) VC TOP1-0.0167 (-0.2795) SIZE 0.0157 *** 0.0170 *** 0.0151 *** (2.7074) (2.9371) (2.5706) LEV -0.1371 *** -0.1378 *** -0.1368 *** (-5.5974) (-5.5984) (-5.5852) DUAL 0.0280 *** 0.0313 *** 0.0293 *** (4.1464) (4.5926) (4.2848) IND YES YES YES YEAR YES YES YES Adj-R 2 0.1575 0.1489 0.1583 F Value 5.7478 *** 5.4434 *** 5.2983 *** **Significant at the 0.05 level, ***Significant at the 0.01 level. The values in the brackets are t values In this paper, we use the China s GEM listed companies from 2010 to 2014 as the research sample to investigate the impact of VC and ownership concentration and their cross terms on enterprise R&D investment. The empirical results show that: R&D investment in VC-backed companies is more than non-vc-backed ones. There is a significant negative correlation between ownership concentration and R&D investment, that is, with the increase of ownership concentration, the investment of R &D becomes less. VC has not yet played a significant role in the impact of ownership concentration on R&D investment. These findings indicate that the GEM listed companies should introduce VC, in order to promote enterprises to devote more R&D investment and enhance the R&D capability of enterprises. Enterprises should avoid high degree of concentrated ownership, so as to prevent the adverse effect of ownership concentration on R&D investment. Moreover, the conclusions of the study have some implications for the further understanding of the R&D behavior of enterprises. References [1] Wang N, Lv YJ. Influence of ownership structure on R&D investment and time lag analysis Evidence from high-tech enterprises. J Science and Technology Management Research 2012; 14: 140-143. [2] Zhang XY, Liao L. VCs backgrounds, IPO under-pricing and post-ipo performance. J Economic Research 2011; 6: 118-132. [3] Large D, Muegge S. Venture capitalists non-financial value-added: An evaluation of the evidence and implications for research. J Venture Capital 2008; 10(1), 21-53. [4] Mai YY, Li JT, Xiong C. Impacts of venture capital and angel investment on the innovative activity of venture enterprises. J R&D Management 2012; 24(2): 79-84. [5] Cai D, Chen ZL, Liu XP. Impacts of venture capital on R&D activities of Chinese entrepreneurial firms. J R&D Management 2015; 27(5): 1-11. [6] Xu H, Wan DF, Xu J. Venture capital s background, shareholding ratio and start-up firm s R&D input. J Studies in Science of Science

Hongxing Wen and Kui Xia / Procedia Computer Science 91 ( 2016 ) 519 525 525 2015; 33(10): 1547-1554. [7] Kortum S, Lerner J. Assessing the contribution of venture capital to innovation. Rand J Journal of Economics 2000; 31: 674-69. [8] Chesbrough H. Managing open innovation: chess and poker. J Research Technology Management 2004; 47(1): 23-26. [9] Gou YN, Dong J. Influence of venture capital s background on firms technical innovation input. J Science Research Management 2014; 35(2): 35-42. [10] Xie WP. Venture capital, R&D investment and enterprise value Empirical data from Chinese GEM listed companies. J The Chinese Certified Public Accountant 2013; 5:47-52 [11] Yafeh Y, Yosha O. Large shareholders and banks: Who monitors and how? J The Economic Journal 2013; 113: 128-146. [12] Liu SQ, Liu X. Effect of ownership structure on corporate R&D investment Based on data of listed manufacturing corporations from 2002 to 2008. J Soft Science 2010; 24(7): 32-36. [13] Feng GF, Wen J. An empirical study on relationship between corporate governance and technical innovation of Chinese listed companies. Journal of China Industrial Economics 2008; 7:91-101. [14] Yang JJ, Sheng S. An empirical study of the effects of corporate ownership structure on technological innovation input. J Studies in Science of Science 2007; 25(4): 787-792. [15] Luo ZY et al. Study on the conduction effect of private enterprises' stockholder's rights structure upon R&D investment behaviour. J Chinese Soft Science 2014; 3: 167-176. [16] Choi SB et al. Ownership and firm innovation in a transition economy: Evidence from China. J Research Policy 2011; 40: 441-452. [17] Ortega-Argilés R et al. Ownership structure and innovation: is there a real link? J Annals of Regional Science 2005; 39: 637-662. [18] Xiong B, Chen L. Venture capital and governance structure of high-tech enterprise. J Contemporary Finance & Economics 2006; 4: 65-69. [19] Wu CP et al. The role of venture capital in the investment and financing behaviour of listed companies: Evidence from China. J Economic Research 2012; 1: 105-119. [20] Fu LM et al. VC is a more positive investor? Evidence from the innovative investment of GEM listed companies. J Financial Research 2012; 10: 125-138. [21] Wang HJ et al. Investment of private equity and the policy of cash dividend. Journal of Accounting Research 2014; 10: 51-58. [22] Yang B, Yuan TR. Research on the relationship between venture capital involvement, R&D investment and innovation performance. J Science & Technology Progress and Policy 2013; 30(13): 83-86.