Nội dung

Le Thi Ngoc Bich et. al. / Journal of Economic Development, 24(3), 45-65 45 Which are determinants of firm innovation in Vietnam? A micro analysis LE THI NGOC BICH Post and Telecommunication Institute of Technology – bichltn@ptit.edu.vn VU TRONG PHONG Post and Telecommunication Institute of Technology LE THI NGOC DIEP Post and Telecommunication Institute of Technology ARTICLE INFO ABSTRACT Article history: This study sets out to investigate the factors influencing Vietnam firms’ innovation in various sectors by using World Bank (2015) enterprise survey of 996 firms across the country. We employ ordinary least squares (OLS), probit model, and marginal effect to estimate the impact of firm characteristics, industry characteristics, and business climate on different facets of innovation, including technology and non- technology. Quantitatively, we find that direct exporters, firm size, state ownership, email using, and competition increase the probability of technology innovation. Meanwhile, foreign ownership impacts negatively on innovation in all aspects, technology and nontechnology innovation. Firm age and bribery are not influential factors to innovation in all cases. From the findings of analysis, a few policy implications regarding the studied factors are drawn for better environment for firm innovation. Received: Nov., 14, 2016 Received in revised form: June, 26, 2017 Accepted: June, 30, 2017 Keywords: Determinants of innovation Direct export Foreign ownership Technology innovation Non-technology innovation 46 Le Thi Ngoc Bich et. al. / Journal of Economic Development 24(3), 45-65 1. Introduction Vietnam, according to World Bank, is assessed as a development success story. After the reforms launched in 1986, Vietnam has made remarkable progress and transformed from one of the poorest countries to a lower middle income country with per capita income of $1960 by the end of 2013. Vietnam’s growth rate has been around 6.4% per year on average for the last decade. In addition, the country has been successful in reducing poverty, and the people living in poverty decreased from approximately 60% in 1990s to below 10% recently. However, the economic growth remains moderate and below its potentials, relying mostly on physical capital, natural resources, and cheap labor. The power of these sources is diminishing while Vietnam is likely to face the so-called middle income trap. To boost its economy and develop sustainably it is time for Vietnam to make innovation to become the drive of productivity gains, especially when the nation is facing fierce competition in globalizing markets. Certain innovative improvements have recently been reflected, yet it still lagged far behind developed countries. According to Global Innovation Index (GII), which is annually co-published by the World Intellectual Property Organization, USbased Cornell University and France–based INSEAD Business School, the country ranked 71st and 76th out of 141 countries in 2013 and 2014, respectively. In 2015, Vietnam was among a group of countries that upgraded their innovation performance ranking compared to that of 2014, stood at the 52nd place out of 141 economies worldwide, and improved 19 places from 2014. The improvement in GII can be a good sign for the upgraded innovation in Vietnam; nevertheless, it reflects only a part of the whole picture of the Vietnamese situation. It is undeniable that innovation in both private and public sectors in Vietnam has lately emerged and there is still a lot of room for improvement. Capability of innovation is weak and the national innovation system is uncoordinated and fragmented. In the business sector, research and development is not properly noticed and faces resistant obstacles, while in the public sector, despite specific privileges, it seems to work inefficiently. World Bank (2017), in an analysis of the Vietnam’ science, technology and innovation (STI) system, highlighted strengths and weaknesses of the country. Accordingly, there are some advantages for STI such as strong economic performance, geographical location, sizeable labor force, or certain achievement in basic education. However, like many other developing countries, there are still many existing problems deterring Vietnam from the development of STI. The resistant weaknesses include infrastructure deficiencies, inefficient education system, limited access to finance for enterprises, and inadequate STI government arrangements and policy implementation. To have further understanding of these strengths and weaknesses this study aims to Le Thi Ngoc Bich et. al. / Journal of Economic Development, 24(3), 45-65 empirically investigate the factors influencing innovative activities of Vietnamese enterprises from different aspects, technological and nontechnological innovation. Using a firm-level data set, the study is intended to draw insights into the deterrents of innovation in order to draw possible suggestions for policies. It will point out the impacts of each element on innovation using empirical evidence, which will be persuasive clues for further implication to help government and other stakeholders perceive where to target their efforts in an attempt to provide favorable conditions for innovation. The remainder of the paper is structured as follows. Section 2 reviews literature on innovation. Section 3 describes methodology and data used in the study. While Section 4 presents the findings and discusses the results, Section 5 concludes the paper and provides some implications. 2. Literature review Different views have been held on measurement of innovation and innovation determinants. Generally, innovation is still an ambiguous concept with different definitions, and there are many controversial opinions on its determinants. Schumpeter (1976) shaped the theoretical framework for innovation, categorizing innovation into five types: (i) launch of a new product or a new species of existing product; (ii) application of new methods in production or sales of a product; (iii) opening of a new market (the market for which a branch of the industry was not yet 47 represented); (iv) acquiring of new sources of supply of raw material or semi-finished goods; and (v) new industry structure such as the creation or destruction of a monopoly position. He claimed that there is a trade-off between innovation and market power of large firms. In other words, to have a rapid technology progress we must be willing to accept imperfectly competitive markets for the reason that in perfect competitive market, where firms produce and sell the same products, there is no incentive to innovate. In contrast, innovative activity is more likely to be favored by large firms and high concentration in imperfectly competitive markets. To examine Schumpeter’ hypothesis, Symeonidis (1996) reviewed many empirical works on the relationships among innovation, market structure, and firm size. The idea that market power and large firms stimulate innovation was found inconsistent. Precisely, this positive relationship can occur when certain conditions are met, such as sunk cost per individual project and economies of scale and scope in the production of innovation rent. Hansen (1992) used the proportion of the sales from new products and total sales as the indicator of innovation and found that both firm size and firm age tend to be inversely related to innovative output. In order to measure the correlation between corporate ownership structure and innovation, Francis and Smith (1995) employed empirical techniques, indicating that diffusedly held firms are less innovative than firms with higher ownership concentration. In other words, concentrated 48 Le Thi Ngoc Bich et. al. / Journal of Economic Development 24(3), 45-65 ownership and shareholder monitoring are effective at lessening the high agency and contracting cost associated with innovation. Defining innovation as activities related to improvement of production and/or process, Lee (2004) examined the linkage that innovation has with characteristics of firms and industries in Malaysian manufacturing sector. The findings suggested that firm size is positively related to innovation because large firms have more chances to access substantial resources and have greater capacity to innovate. Furthermore, ownership structure also impacts innovative activity due to its determination on financial resource through equity market, while sole proprietorship firms are less innovative than private limited and public limited firms. In a study on innovative activity of small- and mediumsized Australian manufacturing businesses, Bhattacharya and Bloch (2004) found that size, R&D intensity, market structure, and trade shares are productive of further innovative activity for the full sample and high-tech enterprises while fewer variables are significant for low-tech ones. Wan et al. (2005) employed data of 71 companies in Singapore, investigating innovation in a more complex and broader context as a process of generation, adoption, and implementation of new ideas or practices. The findings show the positive linkage between innovation and five elements, namely decentralized structure, presence of organizational resources, belief in importance of innovation, willingness to take risks, and willingness to exchange ideas. Almeida and Fernandes (2007) studied the correlation between openness and technological innovation by employing firm-level data in developing countries. They considered technological innovation in terms of whether firms introduced new technology that substantially improved production of its main product in the last three years to the surveyed time. The results showed that firms involving in international trade, export and import, are more likely to adopt new technology. Moreover, it was found that majority foreign-owned firms tend to be more involved in innovative activity than minority foreign-owned firms or domestic firms. Unlike Almeida and Fernandes (2007), this study detects the less innovative tendency in exporting firms, explained by the overwhelming presence of firms with no export in data. Divided innovations of small- and medium-sized enterprises in a low-tech sector (food and beverage) into two aspects: green and non-green innovation, Cuerva et al. (2014) analyzed the differences between factors influencing these two kinds of innovation. The results indicated that technological capabilities such as R&D and human capital are drivers for the conventional innovation, but not the green innovation. Regarding the effects of FDI on the innovative performance of domestic manufacturing firms in India, Khachoo and Sharma (2016) revealed that FDI has a moderate impact on innovative activity of firms residing in identical industries. About the linkage between governance and innovation, precisely the effect of Le Thi Ngoc Bich et. al. / Journal of Economic Development, 24(3), 45-65 corruption on innovation, Veracierto (2008) indicated that under certain parameter ranges, small increases in the penalties to corruption likely result in large increases in product innovation. Remarkably, Aghion et al. (2005) found strong empirical evidence of an inverted Ushaped relationship between product market competition and innovation, implying that competition discourages laggard firms from innovating while encourages neck-and-neck ones to innovate. For the case of Vietnam, the studies on innovation are still limited in quantity and detailed analysis. Nguyen et al. (2013) gave a diagnostic overall review on national innovation systems, analyzing strengths and weaknesses of the institutions, and policies and linkages that characterize the country’s national innovation systems. When it comes to empirical research, there are few papers providing insights into innovative activities of Vietnamese enterprises as a whole and determinants of innovation in particular. Nguyen et al. (2008) empirically examined different aspects of innovation and argued that they are major determinants innovation of exports by Vietnamese SMEs. Employing data from small and medium manufacturing enterprises in Vietnam, Nguyen et al. (2016) demonstrated a positive linkage between corruption and innovation. Precisely, informal payments by Vietnamese firms are shown to foster overall innovation and product innovation. Due to the lack of statistic investigation for Vietnam, this paper would be one of studies taking initiatives in gathering empirical evidence on innovation of Vietnamese firms, which is 49 likely to imply meaningful implications for government and enterprises to positively act and change the situation. In order to analyze innovation in various facets, we take a clear and broad view of World Bank (2004) as the main reference on the understanding of innovation. It suggested that innovation should cover not only “technological innovation,” which is defined as the diffusion of new products and services, but also non-technological forms of innovation. The latter is defined as the introduction of new management or marketing techniques, the adoption of new supply or logistic arrangements, or improved approaches to internal or external communication and positions. Accordingly, this study will examine various aspects of innovation, namely: (i) whether firms have new or significantly improved products or services; (ii) whether firms have new or significant improved method of manufacturing or offering services; and (iii) whether firms have new organizational structure or management practice. Regarding innovation atmosphere in developing countries like Vietnam, the aforementioned study pointed out that firms are deterred from innovation by weaknesses detected from three important elements, including levels of educational attainment, business environment, and infrastructure. Different phases of industrialization require different educational needs, from basic literacy to tertiary education, and these economies fail in matching education and labor demand. The quality of business environment can be measured by governance conditions, values, and cultural 50 Le Thi Ngoc Bich et. al. / Journal of Economic Development 24(3), 45-65 specificities, which can cause obstacles for business operation in these countries. Finally, the issue of infrastructure in developing world relates to the troubles in telephone infrastructure, transport infrastructure, and other primary components such as sanitation, water, or electricity. These common deterrents, nevertheless, seem to have been neglected in previous studies, probably due to the fact that they are not problems for operation of enterprises in those countries. In an attempt to deal with the listed shortcomings in previous studies and depict precisely the case of Vietnam, this study will capture not only the impacts of conventional elements on firm and industry characteristics, but also innovation climate factors which are highly likely to be obstacles for firms’ innovative activities in the three mentioned aspects. Our study takes on various problems of this issue, and hence represents a precise analysis as well as implying proper suggestions to related actors. 3. Methodology and data 3.1. Empirical strategy This study uses quantitative techniques to examine the World Bank Enterprise Survey (2015) for Vietnam using the Stata software. The empirical estimation employs probit model with marginal effect in order to analyze the factors that may influence firms’ engagement in innovative activities with the assumption on the normal distribution of error terms. The dependent variable for innovation is binary, equal to 1 if firms innovate and 0 otherwise. As mentioned in Section 2, innovation is considered from three perspectives in respective models: new or significantly improved product or service (Model 1), new or significantly improved method of manufacturing products or offering services (Model 2), and new or significantly improved organizational structures or management practices (Model 3). Precisely, in Model 1 for the aspect of innovation in products/services, the dependent variable for innovation is equal to 1 if firms have new or significantly improved products or services in the last three years and equal to 0 otherwise. Similarly, in Model 2 which considers innovation as improvement in method or process, the dependent variable is equal to 1 if firms have new or significantly improved method of manufacturing or offering of services in the last three years. Finally, in examining innovation in terms of changes in organization or management in Model 3, the dependent value for firms having new or significantly improved organizational structures or management practices is equal to 1 and 0 otherwise. Explanatory variables are divided into three groups: firm characteristics, industry features, and business climate of country, equivalent to three estimation steps for each measurement of technological and nontechnological innovation. The propensity of innovation in the three aspects is explained by independent variables indicating firm characteristics in the first step, supplementary industry features in the 51 Le Thi Ngoc Bich et. al. / Journal of Economic Development, 24(3), 45-65 second, and business climate in the third. Following the findings from previous papers on the relation of openness and innovative activity, this study considers the difference in the innovation pattern between firms engaging in direct export and their counterparts, which are firms selling products domestically or exporting indirectly. The variable of openness is equal to 1 if firm exports directly and 0 otherwise. The results will be checked on the common expectation that the firm exporting directly is more likely to be engaged in innovative activity than the other. Firm age, presented by the number of year firms operated up to 2015, is deemed to be an element affecting innovation since the operating time may influence them in many ways such as competence of employees, managerial skills of managers, or relation with government officials. Firms existing longer may have better conditions for innovation, but it could be another way around if new entrants tend to be more creative to penetrate their market. Another element is firm size which is added to see the different pattern in innovation of small, medium, and large firms. Those having less than 20 employees are classified as small firms while medium firms are those having from 20 to 99 employees and firms with 100 employees or more are seen as large ones. The number of workers can reflect human resource of firms and potentially cause specific patterns in organization or management of firms. Small and large firms may have no difference in technological innovative activities, but the difference in the number of employees requires different patterns technological innovation. in non- Additionally, this study formulates three models aiming to consider firms’ foreign ownership since the involvement of oversea investors is more likely to cause certain advantages in technology and availability of physical capital, compared to domestic firms. More precisely, the former tends to have financial source and up-to-date technology from outside border, which is favorable for innovative performance in comparison with the later. In the survey firms reported the percentage of capital owned by foreign privates, organizations, or companies, and concrete values would be used to show the difference in propensity of innovation for each percent increase of foreign capital. In addition, government ownership is also adopted in the model for the reason that state-owned companies in Vietnam may have more privileges in finance or legal procedures than private ones. Like foreign ownership, firms were asked the percentage of capital owned by state organization, and concrete values of state owned capital are employed to see the discrepancy in innovation probability in company with one percent change of government-owned physical capital. The final element of firm characteristics which should be taken into account is the role of internet in firms’ innovative activity. Normally, firms employing internet in their operation are more likely to be active and innovative than others that do not. To highlight the effect of internet on innovative activity, the model will compare the 52 Le Thi Ngoc Bich et. al. / Journal of Economic Development 24(3), 45-65 innovative pattern of firms using email and the counterparts. Similarly, internet users are expected to be engaged in innovative activity more than non-users. The variable for email using is represented by dummy variable, equaling 1 if firms utilize email and 0 otherwise. The coefficient is expected to be positive, implying advantages of internet to innovative activity. In the second step, independent variables of industry characteristics are added to the model together with firm characteristics, represented by dummy variables for group of industries in which firms are operating since different industries have different features in technology and innovation. For example, low technology sectors such as food or textiles are less likely to innovate than high-technology ones such as machinery because the later has more sophisticated products and needs continuous improvement to compete in the market. Nevertheless, the opposite tendency can be true that low technology industries are more likely to innovate since their unsophisticated products such as flavor of foods or design of textile products may be easier to improve. The answer for the difference of industries will be investigated among three groups of sectors, namely low technology sectors, medium technology sectors, and services which are classified based on R&D intensities of OECD Directorate for Science, Technology and Industry (2011). Accordingly, low technology sector include firms operating in the industries of food, textiles and garments, and wood and furniture. Medium technology sector includes firms belonging to those of machinery and chemicals, metal, and some of wood and furniture. Service sectors are the remaining industries, including those of construction, sales, hospitality, transport, etc. Moreover, when considering industry characteristics, this study intends to investigate the role of market competition in boosting innovation of firms with the hypothesis that firms will have more incentives to innovate when they have to compete with others. Due to the lack of data for measuring the degree of competition, this study employs available information from a survey in which firms were asked whether they competed against unregistered or informal firms. The variable value is 1 for “yes” answers, and 0 for “no”. The positive value of the estimator implies the advantageous role of competition in innovation. In the third step, to find out the impact of business climate on innovation the variable relating to governance is added to the model. In fact, weak governance, especially beaucratical system and legal regulations, causes many obstacles for Vietnam enterprises. Corruption can be used as a measurement for this weakness due to the fact that when governance is inefficient, firms are more likely to be forced to pay bribe to get things done. In the survey firms were required to evaluate subjectively how many obstacles caused by corruption they have, using scales from 0 to 4, equivalent respectively to no obstacle, minor obstacle, moderate obstacle, major obstacle, and very severe obstacle. In order to compare the difference in innovation of firms facing Le Thi Ngoc Bich et. al. / Journal of Economic Development, 24(3), 45-65 obstacles in bribery and those without, a dummy variable for corruption will be used in the model, equaling 1 if firms have any obstacle from minor to very severe scales and 0 if firms report no obstacle. It is expected that the firm revealing certain obstacles with corruption is likely to have less innovative activity, i.e. the variable for corruption is anticipated to be negative and statistically significant. In short, the equation used in the empirical study for firm i in sector j is depicted as: Step 1: Innovij = Xij’β + εij where: Innovij is a dummy variable to measure innovation of firms in three aspects equivalent to three models: new or significantly improved product or services (Model 1), new or significantly improved method of manufacturing or offering services (Model 2), and new or significantly improved organizational structure or management practices (Model 3). X’ is the vector of independent variables representing firm characteristics, including export, foreign factor, state ownership factor, firm age, firm size, and the use of email. εij is the error term assumed to be distributed normally with mean zero and constant variance. Step 2: Innovij = Xij’β + I’j+ εij 53 where I’j is the vector of variables representing industry characteristics, namely industry dummy variable and competition dummy variable. Step 3: Innovij = Xij’β + I’j+ I’c + εij where I’c is the vector of elements relating to innovation climate of the country, and to be specific, corruption. 3.2. Data The study uses the data of Vietnam obtained from the World Bank’ Enterprise Survey (2015). The World Bank’s Enterprise Surveys (ES) have collected data from key manufacturing and service sectors in every region of the world for many years. The Surveys use standardized survey instruments and a uniform sampling methodology to minimize measurement error and to yield data that are comparable across the world’s economies. The questionnaire was divided into two parts: (i) seven sections covering firm characteristics in business and investment climates such as sales and supplies, infrastructure and services, degree of competition, business government relations, and investment climate constraints; and (ii) three sections dealing with facts and figures regarding finance, labor, and productivity. Additionally, information on capacity such as use of production capacity and hours of operation was surveyed in manufacturing enterprises. 54 Le Thi Ngoc Bich et. al. / Journal of Economic Development 24(3), 45-65 Table 1 Variables and summary statistic description Variable Obs. Mean Std. Dev. Min Max Newproduct 988 0.3076923 0.4617722 0 1 Newmethod 989 0.322548 0.4676879 0 1 Newmanagement 991 0.308779 0.462223 0 1 Directexport 989 12.0546 28.55432 0 100 Medium 996 0.3453815 0.4757314 0 1 Large 996 0.2640562 0.4410508 0 1 Foreign 994 7.10664 24.8409 0 100 Government 995 1.532663 8.878211 0 99 Age 993 12.75629 9.676159 1 113 Email 991 0.9323915 0.2511996 0 1 Medium-technology 996 0.3624498 0.4809493 0 1 Service 996 0.2720884 0.4452587 0 1 Competition 962 0.4656965 0.4990813 0 1 Corruption 996 0.62751 0.4837108 0 1 Poweroutage 986 0.321501 0.4672896 0 1 The 2015 data are the most up-to-date collection attracting participation of 996 enterprises from various industries in Vietnam. The survey was implemented between November 2014 and April 2016 with a more improved questionnaire than the ones used in 2009 and 2011. The number of observations of some variables used in this study may be less than the total sample due to insufficient data for some enterprises (Table 1). The numbers of firms by key background characteristics are generated for qualitative analysis in an attempt to draw insights into the interaction between different characteristics (Table 2). Table 2 shows that among the firms reporting on innovation from the three aspects, there are 304 firms that changed their products/services (equivalent to approximately 30.8% of the total number of firms), 319 that improved their method of manufacturing or offering