Determinants of product innovation activity, the case of vietnamese SME firms

Five factors are figured out to have significant influence on product innovation activity of firms including firm size, age of entrepreneur, export, competition level and network.. Due t

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

HO CHI MINH CITY THE HAGUE

VIETNAM THE NETHERLAND

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

Determinants of Product Innovation Activity:

The case of Vietnamese SME Firms

BY

DANG THUY TRANG

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, May 2015

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

HO CHI MINH CITY THE HAGUE

VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

Determinants of Product Innovation Activity: The

case of Vietnamese SME Firms

A thesis submitted in partial fulfillment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

ABSTRACT

This study examine two manors of product innovation including improving existing product and making new product with the scope of Vietnamese SME's firms The aim of this study is to figure out what determinants have significant impact on each type of product innovation and whether correlation between these two manors exists Determinants are categorized into internal and external factors The first group includes age, experience of entrepreneur, firm size, training cost, and R&D cost The second group includes competition level, outsourcing cost, external consultant, networking, knowledge

of market/marketing cost, and export orientation Bivariate-probit model is used for attaining research's purposes above The model is applied on 1418 SME's firms which are separated into five sections: textile & shoes, food manufacturing and processing, mechanics and electricity, wood processing and all other sections Five factors are figured out to have significant influence on product innovation activity of firms including firm size, age of entrepreneur, export, competition level and network Marginal effect calculated from bivariate-probit regression are helpful evidence for firms to adjust significant determinant to improve their innovation performance

ABBREVIATION

R&D Research and Development

SMEs Small and Medium Enterprises

CIEM Central Institute for Economics Management

OECD Organization for Economic Co-operation and Development

MIT Ministry of Industry and Trade

CONTENTS

I INTRODUCTION 7

I.1 Problem statement 7

I.2 Research objectives: 8

I.3 Research questions 8

I.4 Scope of the study 8

I.5 Structure of the study 8

CHAPTER II: LITERATURE REVIEW 10

II.1 Innovation 10

II.1.1Definition of innovation 10

II.1.2 Popular indicators of innovation activity in SMEs: 10

II.1.3 Classification of innovation: 12

II.1.4 Characteristics of innovation activities in developing country 12

II.1.5 Comparison between large enterprises and SMEs based on indicator of product innovation 13

II.2 Product innovation 14

II.2.1 Definition of product innovation 14

II.2.2 Classification of product innovation 14

II.3 Reviews of Related Theories 15

II.4 Reviews of Empirical Studies 17

II.4.1 Determinants of product innovation 17

II.4.2 Empirical review of methodology 30

CHAPTER III: RESEARCH METHODOLOGY AND DATA 36

III.1 Data and sample 36

III.2 Variables and Measurement 36

III.2.1 Dependent variable 36

III.2.2 Independent variables 37

III.3 Analytical approach 41

III.3.1 Empirical model, estimation method 41

CHAPTER IV EMPIRICAL RESULT 45

IV.1 Innovation activity of SMEs in Vietnam 45

IV.2 Data description 46

IV 3 Descriptive statistics 47

IV.4 Regression Results 50

IV.5 Marginal effects: 55

CHAPTER V: CONCLUSION AND RECOMMENDATION 58

V.3 Recommendation 59

V.4 Limitation and Future Research 60

REFERENCE 61

LIST OF TABLES

Table 1: The innovativeness indicators for SMEs in different dimentions 11

Table 2: Comparison between large enterprises and SMEs based on indicator of product innovation 13

Table 3: Dependent variables 37

Table 4: Determinants of Product Innovation 40

Table 5 : Innovation Rates in Manufacturing SMEs 45

Table 6: Number of observations with improved and new product 46

Table 7: Four cases of combination between improving existing product and creating new product 46

Table 8 : Descriptive statistics 47

Table 9 : Correlation between some independent variables 49

Table 10 : Summary of results from bivariate - probit regression 50

Table 11: Marginal effect 55

LIST OF FIGURES Figure 1: Conceptual Framework 35

Figure 2: Methodology model 44

I INTRODUCTION

I.1 Problem statement

International economic integration allows more firms to be established and more foreign firms to enter domestics market Under such intensive competition pressure, firms need to continuously innovate their products to attain comparative advantage from other opponents This is due to limited life cycle of individual product and more arising demand of customers M Banbury and Mitchell (1995) proved that the more often firms create new products, the better they perform and the higher their probability of surviving

in long term is This is applicable for both small and large firms (Vermeulena, De Jong,

& O'shaughnessyc, 2005) Chris Freeman and Soete (1997) - even asserted that ‘not to innovate is to die’ Due to such important role of innovation, a large amount of studies have been done to identify factors having influence on product innovation of small firms (Martinez-Ros,1999; Jong, 2006; Fritz,1989; Vega-Jurado, Gutierrez-Gracia, Fernandez-de-Lucio & es-Henriquez, 2008; Hadjimanolis,2000; Freel, 2000)

It requires more than expenditure on R&D for attaining improved or new products Factors with considerable influence on innovation activity that should be taken into account include firm size, production technology ( ratio of sales over fix assets), origin of ownership, spending on market insight (Avermaetea, et al., 2004) How much these factors may affect innovation activity is not equal and depends on economic section of firms Thus, a model simulating impacts of aforementioned elements is necessary to be established

While there are a range of international researches addressing this topic, number

of papers on such matter at Vietnamese scale is quite limited As Vietnam is a developing country, a moderate technical progress may result in huge increase of firm' sale or market share (Hadjimanolis, 2000) Thus, a model applying this matter for Vietnamese scenario is essential to be devised This paper is expected to achieve that target

Both Vietnamese enterprises and policy makers can utilize results of this paper First of all, enterprises can refer to this model when they are in need of promoting innovation for their product As the model points out specific elements and the level such elements need controlled This is also a useful guidance for policy makers In detail, being aware of influential elements of each industry allow policy makers to make proper decision to assist innovation of that specific section

I.2 Research objectives:

The study is expected to figure out determinants with significant impact on product innovation activity of Vietnamese SMEs and influential level of each factors

By understanding the interaction between the internal and externals factors having impact

on product innovation, enterprise’s leaders may adjust firm’s strategy to adapt to characteristics of environment to gain the highest performance in product innovation activity

I.3 Research questions

There are two main questions addressed in this study The first one is what indicators have considerable impact on product innovation behavior of firms The second one is to which extent product innovation is influenced by these factors

I.4 Scope of the study

The study examines Small and Medium enterprises across Vietnam with data extracted from SMEs survey 2011

I.5 Structure of the study

This paper includes five chapters The first chapter - Introduction presents the issue and background of the issue addressed in this paper, then discussing research objective Chapter II–Literature review presents the theoretical and empirical foundation

of the topic which aims at creating a conceptual framework for the research This part explains the definition of innovation, product innovation, theoretical and empirical theory about determinants of product innovation activity of firms, the methods applied to

measure product innovation Chapter III – Research Methodology focuses on the method, data description and the analytical model applied in this paper Chapter IV– Findings and Discussions discuss the regression results and marginal effect of each factor on product innovation probability The final part Chapter V–Conclusion and Recommendations suggest policies for policy makers and for top management of firms when they want to increase the rate of innovating product

CHAPTER II: LITERATURE REVIEW II.1 Innovation

II.1.1Definition of innovation

According to Hyvarinen (1990), innovation activities involve both internal and external activities of firms with the target of creating new products, enhancing existing products, process, governance or marketing Most of innovation activities originate from research and development (R&D) Some may be acquired by firms via other sources such as licensing, seminars, consultants, customers, providers Technologies can be applied in various ways in each firms When technology is aware at a broader sense rather than relating to products only, innovation activities will play a key role in

development of SMEs (OECD, 1982)

II.1.2 Popular indicators of innovation activity in SMEs:

There are three popular proxies usually used to indicate levels of innovation: input, output and impact These indicators could be analyzed in general or particular meaning and measured by quantity or quality The measurement units used most frequently are number and rate Among various inputs of producing, time, capital and labor are the common factors used to measure effectiveness of innovation Because these factors can be collected more easily and exactly than others belong to output or impact, the input method is preferred than the other two methods The second indicator- output involves in analyzing straight outcome of innovation, creation, expertise, proficiency and absorption of technique The typical proxy of output is patent as its data can be found in most country’s statistic system However, patent is not the most significant in all companies and all industries Some innovations may not be registered for patents Besides, patent refers to invention rather than innovation Output method sometimes utilized number of license as indicator The last and the latest among three methods is impact indicators This method relates to innovation in larger aspect and refer to qualitative outcomes rather than quantity Impact is comparison between cause and effect

of innovation and the variation of sales, capital, productivity, development, etc of firm

( Harrison& Hart, 1987;Kamien & Schwartz, 1975; Meyer-Krahmer,1984; Walsh, 1984

and Scholz ,1988)

The table below describe three indicators of innovation when they are examined based on four dimensions including technology, individual, enterprise, market/ environment

Table 1: The innovativeness indicators for SMEs in different dimentions

Technology * Personnel participating R&D

Publications Exports

Success Improved value added Equity increases Enlarged-market Better image Rewards Development of enterprise Internationalization

Products Innovations Markets

Development of branches New enterprises

Employment Regional development Facts

Know-how Economic growths Diffusion of innovation

Source: Hyvarinen (1990)

II.1.3 Classification of innovation:

There are four criteria which are usually used to classify innovation: time, influencer (firm/entrepreneur), market, and technology Firstly, innovation can be ranked based on its newness The order from the oldest to the latest is basic, applied and incremental innovation The time within the concept of innovativeness of small- and medium sized industrial enterprises gives the opportunity to notify the following: degree

of newness of innovation, delay in the reactions of SME to changes in environment and

in the information about them, time-related development of a SME , time dependent innovation process, the delay between the output and impact and input of innovation activities When influencer or causer of innovation is analyzed, the typical classification

is innovation caused by entrepreneur or caused by firm When it comes to innovation in different departments of firms, there are five types of innovation: product, process, marketing, organizational and social innovations (Hyvarinen, 1990) The final criterion is market This means that an amendment is innovation when it is new to the market

II.1.4 Characteristics of innovation activities in developing country

Typical features of developing countries related to innovation activities are listed as below:

 There’s a shortage of policies facilitating innovation and organizations which should be in charge of managing and supporting technology and innovation Some

of them maybe exist, however their performance may not be mature yet These institutions include high tech industrial zone, investing firms, suppliers of technological material The insufficient innovation system at national level reduce technological chances of the whole economy (Fontes and Coombs, 1997)

 Quality of innovative products which are produced by domestic manufacturers are not trusted by consumers Beside, due to small size, the demand of local market is limited and the demand for innovative product is unimportant (Fontes and Coombs, 1997)

 The economy is dominated by small firms While the number of medium and large firms is quite modest According to the theory of relation between size and

innovation activity, such an industrial structure doesn’t encourage innovation There is no complementing relation between small and large firms as found in developed country (Fontes and Coombs, 1997)

 The cooperation between firms and research institutions, universities is not very tight There’s not much new techniques transferred from these organizations to manufacturing firms (Jones & D., 1996)

 Most of innovation activities result from adapting techniques which are initially created or invented in developed countries Thus, an open economy which allows connections with foreign countries is necessary for stimulating innovation

II.1.5 Comparison between large enterprises and SMEs based on indicator of

product innovation

Innovation activity of firms is differed due to scale which is presented in Table 2 below

Table 2: Comparison between large enterprises and SMEs based on indicator of product innovation

Simple structure, easy to apply innovation like new invention or improved process

documents

Unofficial information plays a key role in some case

Adaptation Disapprove of new idea, just

create basic innovation

Easy to approve to use small, new unique idea

innovation

Radical innovation

Source: Hyvarinen (1990)

II.2 Product innovation

II.2.1 Definition of product innovation

There are different ways to define product innovation It is widely interpreted as the activities of developing and launching to market a product which is new, enhanced or significantly improved In specific, product innovation may range from inventing a new product, improving a product's quality, enhancing product's technical features or adding new parts, input and expecting functions to an established product (Acs & Audretsch, 1988) From another point of view, as long as the product is new to market, it will be considered as an innovation, even when alike products are already exist

II.2.2 Classification of product innovation

Different forms of product innovation could be separated into two categories: radical innovation and incremental innovation The first means creating a totally new product while the second refers to improving existing product Radical innovation involves in introducing new product to market This process includes two simultaneous stages with the first stage relates to developing idea, designing product, technical preparation, and the second stage relates to analyzing market Incremental innovation involves in enhancing established functions, technical value and user friendliness This method of innovation is usually utilized at the declining phase of outcome to lengthen expectancy of product (White, Braczyk, Ghobadian, & Niebuhr, 1988)

Although some people prefer to explain this term by the first meaning, only a certain number of firms attain such level of innovation A paper of Wong (2014) even finds out that over 80 percent of new product become failure and not every successful

launches can maintain a steady growth This is however, not a problem as it has been proved that a regular and steady innovation on existing products is more important than a sudden substantial change in products In fact, many industries across the world tend to favor product improvement than inventing a whole new one since the seventy

II.3 Reviews of Related Theories

Neoclassical Economics is the initial method used by economists and business managers to determine innovation activity In this theory, innovation or technology is defined in basic description and simply measured by production function No relation between firm's characteristics and innovation was figured out so all firms shared the same features in term of technology Schumpeter (1934) is the first author carry out deep analysis in the significant impact of technology on economic growth He is also a pioneer considering relationship between firm's characteristics and innovation His study received

a lot of interest from public and more studies were conducted in term on technological improvement after then In the research, Schumpeter takes two elements (firm's size and market competition) into account and assumes they have direct impact on innovation The finding of this study encourages later research on influence of external factors regardless of industry or public level This study, however, does not discuss in detail about other firm's characteristics or the procedure of creating innovation in firms Thus, more references need to be considered for a comprehensive understanding

Various opinions have been suggested to make clear this idea Some typical theory are namely transaction cost economics (Williamson, 1989), the positive theory of agency (Jensen & Meckling, 1976) the evolutionary theory (Nelson & Winter, 1982) and theory resource-based view the firm (Wernerfelt, 1984)

Transaction cost economics analyses costs incurred from transferring of product from seller to buyer Potential costs can arise at three phrases: contact, contract and control (Noote Boom,1999) Contact costs includes cost of searching for appropriate product of buyer and cost of promoting products to potential customers of seller Contract

is the cost of negotiating the specific terms to fulfill transaction The final phrase control

is the cost of ensuring both parties comply to terms and conditions specified in contract

In general, all these costs originate from three reasons The first is uncertainty which includes fluctuation in both parties' behavior and uncertainty of the environment The second is specific investment which mean capital required to use or operate a product after transferring The last reason is information asymmetries Due to such transaction costs which constrain flow of technology between firms, between researchers and firms,

it is recommended that innovation should be internalized rather than transact through third party organization

The second theory mentioning impact of firm's relationship on innovation activity

is agency theory This theory analyses conflicts between principal/agent or manager/shareholder and accused information asymmetry for such conflict which may lead to inefficiency (Jensen & Meckling, 1979) To minimize information asymmetry, agency theory suggest to constrain open corporation and knowledge/ technology transfer

as these relations may create deviation in interest of manager and shareholder and reduce firm's effectiveness While network is necessary for firms in adapting new technology (Tidd & Bessant, 2009), it is not easy to avoid information asymmetry when adapting external knowledge Finally, under assumptions of agency theory, firms may restrain investment of technology to avoid information problems

The procedure of developing innovation process of firm is studied by evolutionary method This method assumes that technology always changes and such change/evolution

is regular and constant Due to the assumption that every adjustment of technology is accumulative, its development relies on its past Pavitt(1987) The most important contribution of this theory is it emphasizes the difference between firms when it comes to technical capability As this is necessary to classify procedure of innovation (Dosi, Freeman, Nelson, Silverberg, & Soete, 1988)

Another theory analyzing innovation of firms is resource based view which aims

at figuring out firm's key resource to develop comparative advantage According to this theory, innovation is a valuable asset of firm as it is based on information which can be accumulated and enlarged continuously (Wernerfelt, 1984) Innovation is considered as a significant source of advantage for firms as it allows firms to outperform competitors or improve weakness Besides, innovation is valuable and scarce, it does not depreciate with use (Itami & Roehl, 1991) and it takes effort to make transfer of innovation It's not easy

to imitate or substitute an innovation due to limit of time and economy of scale Innovation is a key feature creating comparative advantage for firms However, level of innovation in each firms will be determined by specific characteristics of firms such as capital, human or governance structure Thus, comprehensive analysis of firms' features

is necessary in determining innovation activity of firms

The last model is industrial organization This model assumes that technology has

a linear relationship with information and technology is the connector of science and innovation From such a viewpoint, it is understood that innovation is considerably depend on external determinants and firms are shaped by its actions in the past

II.4 Reviews of Empirical Studies

II.4.1 Determinants of product innovation

Regardless of varied methods applied to examine product innovation, all determinants addressed in those papers can be divided into two main types: internal and external factors (Schumpeter,1942; Martinez-Ros, 1999; Avermaetea, et al., 2004; Fritz, 1989) The first group refers to elements related to enterprise's character such as : size, technical staff, professional staff, investment on R&D, spending on marketing and factors involved in features of entrepreneurs, i.e : age, experience with product before The second group comprises of factors affecting firms from outside such as intensity of competition, comprehensive relationships, outsourced consultants

In another study of Damapour (1991), different factors having influence on innovation activities are divided into three groups as follows:

 Enterprises’ human resource which include owner, manager, technical staff…

 The enterprise’s characteristics

 External factors having interaction or impact on enterprise, such as competition level, network…

In 1990, King extended such theory with two antecedences of innovation which are: national innovation policies and Inter-firm Linkages Yoshihara also mention national economy in his study in 1976 He explains that environmental changes provide opportunity for innovation which include both threats and encouragement for firms' business In specific, enterprise' environment can be separated into two categories: the first one is direct environment including market demand and supply, customer taste, anti-business attitudes as well, the second one is macro factors such as national and international economy, political situation, education, technology, population and so on

In sum up, determinants of innovation activities consisting the argument and measurement in relevant references are summarized as below:

II.4.1.1 Internal factor

a Characteristics of owner/ manager

The roles of owner/manager's characteristics are approved by many studies to have impact on innovative ratio of firms Schumpeter (1934, 1942) is one of the first authors figure out the influence of entrepreneurs on innovation There is more researchers agree about the role of this factor recently (Mascitelli, 2000) Indeed, the owner/manager

is supposed to analyze market and determine appropriate moment of investing to enjoy new advance technique (Fontes & Coombs, 1996) Besides, as entrepreneurs directly

keep in touch with major relations of firms, they will decide whether or not to apply innovative idea from these potential sources (Lipparini, 1994) In small business, the manager/owner is mainly responsible for strategic decision then his role is more significant as compared to large business where the decision is made through various levels of assessment Drucker (1985) and Urban & Hauser (1980) stated that small firms' innovative behavior tend to be influenced by individual features while large firms innovation is determined by characteristics of firms such as product, capital, investment The innovation activities of individuals form an important part of the innovation resources and organizational behavior

The age of the entrepreneur is expected to be negatively related to innovation The reason is younger manager/owner has more incentive to perform innovation as they have

more time bonding with firms (Diederen, van Meijl, & Wolters, 2000) Avermaetea, et

al., (2014), proposed a contrast argument which states that "a degree in science or technology and long-time experience in the firm are thought of as indicators for innovative capabilities"

The accumulation of knowledge and experience is approved by many studies However, whether an experienced manager/owner is more innovative than a younger one

or not hasn’t been empirically certified clearly (Romijn and Albaladejo, 2002) Background and skill of entrepreneur are important factors to be analyzed as well Some authors proposed that a manager/owner who is post graduate will be more eager to innovate than a graduate one This is due to a range of skills such as communication, social connection, technical fluency created during the studying time (Cohen &

Levinthal, 1999) However, there are not many papers recently approving such a

suggestion (Romijn & Albaladejo, 2002) Finally, attitude of entrepreneurs should be considered as a determinants of innovation because firms will enjoy more innovative opportunity when entrepreneurs have interest in advanced technique

b Firm size

There are two completely opposite arguments about influential sign of firm size on innovation Schumpeter (1942) and Frits (1989) agreed with the positive relationship They explained that only large firms with prosperous budget can sponsor innovative activities which usually require remarkable expenditure Large firms which have larger budget to hire/purchase new technique tend to produce more innovative products The negative relation is approved by Fritz (1989) and Martinez-Ros (1999) They stated that small firms with less hierarchy can quickly response to any change of market Besides, while large firms can proceed some options such as take over, merging and acquire to increase comparative advantage, small firms have less options so innovating product is usually resort to Moreover, large firms is often associated with high market share Large control power may lead firms to under evaluate innovation activity

Symeonidis (1996) provide detail arguments for the impact of size on innovation

of firms Firstly, R&D requires huge initial investment, thus firms must have large sales

to make up for investment costs Secondly, economy of scale exists in producing innovative products Thirdly, large corporation with diverse business area tend to better exploit advantage from sudden innovation Fourthly, large firms can distribute risks from

an investment of innovative product via other projects of different industry Fifth, it is easier for large firms to mobilize capital from outside Sixth, market power allows big firms to make use of innovation better, thus these firms are more eager in proceeding innovation

Regarding this factor, Acs and Audretsch (1988) proved the strong influence of size on innovation activity, via both input and output It is even found out that the relationship between size and innovation activity is not linear When size increases, there will be more opportunity innovation will happen Until a certain level, when size continue to increase, innovation rate will reduce (Kamien& Schwats, 1982)

Actual relation between firm size and innovation tested by empirical study is rather diverse In 1982, Kamien and Schwartz checked this relation among different industries Only chemical section presented a significant relationship In another paper of Mansfield in 1981, fundamental studies are mainly carried out by large firms, but small firms invest more for innovating product

Most empirical studies analyze firm size factors by relating innovation activity to size measurement across firms in different industries Firm size is proved or assumed in those studies to be exogenous Innovation has direct influence on firm growth, then on firm size Thus size of enterprise in year t is assumed to be related to innovation activity

in year t-1 (Scherer, 1992)

In case innovation activity has relation with undetermined elements which are serially correlated, firm size is expected to be correlated with those elements in year t Thus, when regress innovation on size, the result will be bias However, some others propose that innovation influences firms through a couple of year In another way, influence of innovation on firm size is lagged across some years Thus, endogeneity is not

a big issue anymore

Besides, there is another issue when comprising firm size into the analysis of innovation That is the potential correlation between firms size and industry - level factors such as technological opportunity which may positively affect innovation This may cause bias result to regression with sample including different industries The solution is adding control variable for such industry effect (Cohen and Levin ,1989) However, adding control variable may not help solve the problem completely as many large firms work on various industries To prevent this problem, it is suggested to categorize industry to deeper level of 2 digit, but this then encounter another problem that each sectors within one industry may have difference in some features

In sum up, firm size is proved by many research to be influential to innovative activity of firms The sign, nature, and the method to analyzed this relation, however is varied across each study

c Human capital and investment in training &marketing

Beside capital, labor force is another factor that makes up production function of firms This mentions important role of human capital on firms' performance In 2004, Avermaetea, et al pointed out that “the skills of the workforce and the firm’s investment

in such skills contribute substantially to product and process innovation in small food firms” Spending on regular training and coaching for staff is a wise investment for firms

to gain more innovation in future ( Freel, 2000)

It's used to be believed that innovation is determined by specialists, professors, or engineer (Tidd, 1997) Recently, innovation has been proved by many studies to be resulted from effort of total workforce of each company (Clark & Fujimoto, 1991) In spite of humble contribution of each person, the accumulated influence of the whole staff

is considerable Employees may involve in innovation via one among below steps: discovering opportunity, generating idea, developing, trying and putting innovation in market (Kleysen & Street, 2001) In general, the role of the whole staff in developing new product is undeniable

Employee involvement is agreed to influence innovation However, how much such contribution is depends on staff's actual knowledge and skill (Warner 1994) Experienced and skillful staff usually require high salary which is equivalent to their efficiency They prefer to work for large firms where financial resource and working condition is better than at small firm (Freel, 2000) This is one direct cause leading to variation in innovation activity between large and small firms Workforce skill even contribute 60% to innovation of service firms Due to above arguments, it is suggested that firms should spend more on staff training to enhance workforce capacity to attain higher innovation rate

e Origin of ownership

Origin of firm ownership is divided into two types: foreign owner and domestic owner For firms situated at developing countries, multi-national companies with professional working style, tremendous physical resources is believed to be more active

in innovating Besides, multi- national companies may receive transfer of technology from foreign subsidiaries Thus, these firms have higher rate of innovating compared with purely domestic firms (Cave,1982 )

In a study of Baldwin & Sabourin (2000), origin of ownership is figured out to be

positively correlated with process innovation activity About half of foreign firms apply process innovation while relevant number for domestic firms is only a third The probability of carrying out process innovation of foreign firms are 50% higher than domestics firms Regarding product innovation, the result presents no different between two types of company

f Intensity of physical capital

The production technology's features may have influence on whether the innovations will be introduced or not Production technology can be verified via intensity

of physical capital "Firms with more capital intensive technologies will tend to innovate more if, as expected, the rents of innovation are less threatened as, to exploit the innovation, high investment in physical capital is required" (Martinez-Ros, 1999) More capital intensive processes may have negative effect on innovation as they are more automated and rigid Thus, total influence of capital intensity is uncertain

g R&D activities

Enhance and improvement in technique are considered to be highly related with innovation in both product and process As research and development activities play a key role in creating technological advance, they usually receive attention from many companies (Grunert, Hartvig Larsen, Madsen, & Baadsgaard, 1996) In food processing sector, R&D activities are disapproved by some authors to have influence on innovation

However, many others agree of a positive relationship between these two elements (Huiban & Bouhsina, 1998)

To measure innovation, many authors utilize the number of patents and spending

on R&D of firms In a study of Crepon, Duguet, & Mairesse (1998), it has been proved that the sale of an innovative product increase 5% when R&D spending rises up 10% Another study of Baldwin & Diverty (1995) which applies the data from a survey of 500 Canadian firms figures that the chance to invent a new product or enhance an existing one will increase 24% for firms which have spending on R &D This study also asserts that R&D and firm size have the biggest influence on innovation Firms not performing R&D have only an 11% opportunity of innovating, while the rate for firms with R&D expenditure is 41%

h Export orientation:

International trade allows countries to focus on their own competitive advantage International trade also requires firms to provide the best product they can produce Especially for developing country, trading with foreign customers allows firms to gain higher margin than domestically trade However, the higher customers pay, the higher quality they require Thus, export tends to encourage firms to perform more innovation

In a research of Higon and Driffield (2007), data shows that about 45% of SMEs business exporting their products carry out innovation while this number for SMEs

trading domestically only is 26% In another study of Matinez Ross (1999), export

orientation is included into the model as control variable This study expect that firms trading in foreign market resort to innovation to gain more competitiveness The vice -versa relation is also anticipated which is explained that firms with high innovation rate may be more keen on export activity as they have sufficient intangible resources to maintain development Data collected from study of Becker and Egger (2013) also supports positive relation of innovation and export In detail, about 62% of export firms carry out either product or process innovation, while only 30% of firms surveyed having

no innovation

i Sector/ Industry characteristics:

A range of conceptual and empirical tests are addressed to clarify the variation of product innovation among different industries In general, there are three main approaches for this issue Firstly, research is proceeded with data including a variety of sectors without making comparison (Bougrain& Haudeville, 2002; Brouwer, 1997; Hadjimanolis, 2000) The first two authors agree the existence of difference among industries but the specific innovative manors are not examined The second author group states that not enough evidence for variation of innovation among sectors is available and they also ignore patterns of innovation Hadjimanolis (2000) separates the data into five sectors, but no comparison are mentioned In sum up, results of these papers maybe bias

as total effect may overwhelm significant factors

The second type of study restricts their sample within one sector Romijn and Albaladejo (2002) includes a lot of elements in their model However, data is constrained

at electronics and software firms only As a result, consequence of this approach is insufficient generalizability

In the third type, differences across sector are analyzed for example, manufacturing versus non-manufacturing, manufacturing versus services and high-tech versus low-tech Acs & Audretsch (1988) include a lot of industries in their study No comparison among sectors are made, but the difference between large and small firms is examined and the variation due to innovative intensity is addressed The analyzing method is the same for researches of Kim et al (1993), Rogers (1971) and Bhattacharya and Bloch (2004) Only a few studies clarify differences in detail Oerlemans et al (1998) figured out several differences in the way firms utilize their resources Freel (2003) proved that variation of innovation behavior across sectors is remarkable while Roger (2004) explains clearly the different of sign and intensity of every factors among manufacturing and non-manufacturing sectors

II.4.1.2 External factors

a Competition level

According to Arrow ( 1962) and Fritz ( 1989), intensity of competition encourage firms to proceed more innovation because renewing product is an effective method to gain more comparative advantage However, other firms may have opposite reaction toward competition Schumpeter ( 1942) stated that severe competition restrain firm's profit, so firms will not eager to carry out innovation in highly competitive environment

In a research of food industry in Canada, Baldwin and Sabourin (1999) pointed out the non- linear relationship between competition and innovation In detail, competition among 0-5 firms or among over 20 firm will restrain firms from innovating while competition among 6-20 firms will encourage innovation The result also figures out that probability of doing product innovation in a moderate competition is 58% , in a low competition is 38% and in high competition environment is 48%

In another study of Bonanno and Haworth (2008), game theory model namely Courtton - Bertrand is applied to analyze firm's selection between product and process innovation The result indicates that the Courton competitor who selects to compete by quantity determines to innovate its product, while Bertrand competitor who competes by quantity decides to innovate its process

Another aspect related to effect of competition intensity on firm's innovation and being examined by several studies is firm size According to Acs and Audretsch (1987), large firms tend to innovate in condense market with less than five firms Meanwhile, small firms prefer to innovate in a competitive market without barrier to entry Blundell

et all (1995) found out negative relation between market share and innovative activity of firms This means that start-ups tend to have more innovation than larger firms

b Outsourcing and exploiting external assistant

Outsourcing - which means authorizing agents to perform a part of work has become a global tendency This method allows firms to reduce cost by exploiting cheap

labor in other country or by focusing resources on their advantage However, outsourcing may reduce opportunity to carry out innovation of firms Martinez-Ros (1999) The agent only performs as guided to avoid discrepancy while firms' staff who have better understanding of the company's target will think of more innovation Rothwell and Dodson (1991) found out the opposite relation in their study In detail, outsourcing allows companies to enhance products without investing large amount on technique or equipment They confirm that small firms' innovative rate is higher when carrying out vertical integration - the opposite of outsourcing The reason is vertical integration allows firms to absorb knowhow and knowledge transferred from their suppliers or contractors The result of their model states that half of innovative firms are integrated with supplier while only 10% closely contact with customer

c External information/ Networking

Small firms rarely have sufficient abilities to proceed innovation by themselves Thus, they usually depend on outside sources of information and input Especially for small firms

in food industry, the capability of achieving knowledge, experience…from external sources plays a key role in innovation activities of firms (De Propris, 2002; Freel, 2000)

Success of innovation has been argued to be related to external network (Rothwell,1991) In fact, small firms which pay attention to this issue and energetic in

connecting with partner tend to perform better (Lybaert, 1998) Common features of

these firms are keen on science, being active in collecting information and close relationship with suppliers Cooper & Kleinschmidt (1995) proves that the more frequent firms contact with partner, the more successful their innovation are Relationship with academic institution, rivals or consultants are certified to play a key role in innovation activity of firms as well Similarly, it is widely agreed that extending network with coworker, suppliers, buyers allows firms to enlarge their innovative opportunity (De Propris; Diederenet al., 2000; Freel, 2000) In conclusion, whether it is internal or

external linkages, firms are recommended to strengthen all of their relations in case they need better innovative performance

Collaboration between firms or inter firms cooperation is emphasized on many study about innovation De Propris (2000) states that "inter-firm linkages make up the

‘missing input’ explaining small firm’s innovation performance" This means cooperation between small firms is very necessary for their innovation The reason is large financial requirement for investment in sophisticated production technique is met when inter-firm linkages exist It is popular in rural area of Vietnam where farmhouses share one truck or one harvester as these equipments increase their efficiency but rather costly It is also pointed out in studies of Bougrain & Haudeville (2002) that shortage of capital and knowledge demand firms to collaborate with each other for higher efficiency and more advanced products Nevertheless, during such cooperation, some firms express worry for the possibility of imitating confidential knowhow or difficulty in controlling the business outcome when sharing power with other firms

d Knowledge of market

Knowledge of market is defined as the awareness of demand of customers and efforts of companies to fulfill such needs This is an important source of external information like research institutes or advisors As firm's success is determined by its ability to meet customer' need, innovation is successful only when it fulfill customer's demand (Twiss, 1992) It has been verified that firms doing market search to understand customer need are more likely to success in innovation and have more advanced products (Cooper & Kleinschmidt, 1995) Moreover, being aware of market demand allows firms

to estimate potential profit of innovated product once produced This is an important foundation for firms to determine whether to innovate or not Some typical features of demand should be examined are: time of demand, scale of demand, demand elasticity

In food industry, study of Grunert, Hartvig Larsen, Madsen,&Baadsgaard (1996)

is considered as a foundation reference which describes a typical analyzing scheme for innovation activity In this study, knowledge of market is suggested to be a major determinant of product innovation beside R&D activities In another study of Steward- Knox & Mitchellin in 2003, market awareness is proved to ensure success of product launching Some articles provide evidence about the fact that marketing activity is a main source of product innovation in food industries In a study of Avermaetea, et al (2004), proxy of market awareness is spending on marketing activities as % of the firm’s turnover This implies that the larger business spend on marketing, the larger amount of market information will be gained

In conclusion, market awareness plays an undeniable role in development of products Doing market research is necessary for firms to adapt to customer need timely It's recorded that improving or creating new product in many cases originate from a market research campaign (Hadjimanolis, 2000; Kim, Kwangsun & Jinjoo, 1993) Thus, firms who intend to develop their product should focus on this element

e Country environment

The relationship between innovation at the firm level with environment of country was first found out by Whitley (2000) The economic mechanism of countries is divided into six categories based on the rate of privatization These six groups include : economic planned, cooperative, collective, deregulated, liberalized economy The deviation of institution is the root of differences in firm’s incentive and campaign to innovate For example in an economic planned, product will be more standardized, and product quality may not be continuously improved like in an liberalized economy The influence of national environment on firm’s innovation maybe explained by the diamond theory of Micheal (1986) as well

II.4.2 Empirical review of methodology

Innovation is a qualitative variable which is usually measured by two options Yes/No Several qualitative regression methods have been applied to analyze relationship between determinants and innovation In a study of Martinez-Ros (1999) and one of Jong (2006) , "innovation activity is measured in terms of a discrete variable that takes the value

of 1 if the firm has innovated in period t, and 0 otherwise” In a study of Vega-Jurado, Gutierrez-Gracia, Fernandez-de-Lucio, & es-Henrıquez (2008), multinomial logistic model

is utilized In this paper, development of products includes three types: the first type refers

to products appear the first time on market; the second type is produced by firm for the first time but already exist on market; the final type is added new features from existing products In another study of Avermaetea, et al.( 2004), multinomial logistic model is applied as well However, based on level of product innovation, dependant variable is measured by four levels : the first group having no new product; the second group has new product but none R&D spending; the third and final group have both new product and R&D spending but the third spends less than 1% of their revenue on R&D while the final spends more than this rate Hadjimanolis (2000) also divides dependant variable into four group and each is measured by a scale of five levels In summary, based on target of study and availability of data, product innovation is measured by specific way

II.4.2.1 Multinomial logit model

The most popular model utilized in evaluating innovation is multinomial logit model This model is applied in a study of Avermaete (2004) regarding determinant of product and process innovation in small food manufacturing firms

A survey with respondents are firm’s manager/ owner is conducted before the study According to the responds, firms are divided into four groups The first group is non-innovation including firms having no new or improved product The second –

“traditional group” refers to firms having no investment on R&D activity, but having improvement for their products/process The last two group “Follower and Leader” both conduct R&D activity and have at least one innovation on their product/ process While

maximum spending of “Follower” on R&D is 1% of their yearly turnover, this counterpart of “leader” is more than 1% In general, innovation level of firms is measure

by a category of four values, which is based on their outcome from innovation activity (regardless of product or process) and their expenditure on R&D

Dependent variable take four discrete values as described above Independent variables are divided into two groups: internal capability and external information The first group consists of entrepreneurial characteristics (age, science qualification, firm experience), skill of the workforce (number of qualified technical staff, number of managerial and professional staff, % qualified technical staff, % managerial and professional staff) and investment in know how ( training cost, marketing cost) The second group includes services outsourced (administrative, marketing and technical consultants) and source of innovation ( similar firms, supplier, customers, contractor)

The functional forms of this study is as follows:

)1.(

&

exp)

N T

N T

&

exp )

N F

N F

&

exp )

N L

N L

&

exp )

T F

T F

&

exp )

F L

F L

&

exp )

T L

T L

During running the regression, non - innovators and leader is by turn selected to be base value The purposes is to compare other value with these two extreme

II.4.2.2 Probit model

Product innovation and process innovation are examined simultaneously Innovation

in general is considered as consequence of interaction between technological capital at the early stage of the period, technological opportunity from market and other factors relating

to features of firms and market In general, the determinants of innovation include:

E( Iit) = f( Git-1,τ,Xit-1)

In which I is innovation in product or process of firms, G is technological capital,

τ represents technological opportunities and X refers to group of variables involving in market and firm's characteristics

Univariate probit model is applied for the analysis Product innovation/ process innovation is alternatively regressed on determinants above When firms carry out innovation, variable product/ process innovation takes value of 1, and 0 for no activity, respectively Nominal variable product/process innovations are measured by the following equations:

it >0, IPROD it =1, otherwise IPROD it =0

In the second equation, when IPROC*

it >0, IPROC it =1, otherwise IPROC it =0 Git-1 is technological capital stock of firm which is a lagged value to explain influence of accumulative capital from the past XFIRM is a group of independent variables related to

characters of firm includes size of firms, production technology, export activity and origin of ownership, vertical integration XMARKET is group of independent variables related to features of market which consists of competitive intensity, demand growth and homogeneity of product Error terms ε it, ϑ it follows standard distribution

In two equations above, capital stock Git-1 can be affected by independent variables XFIRM and XMARKET To solve this endogenous problem, instrumental variable GINST it-1 is added GINST it-1 is regression of technological capital stock on nominal variables involve in industry, time, firm features, and market feature

After running probit model, author raised question that whether unobserved firm effects may have effect on innovation If such variables (for example managerial skill) highly correlated with dependent variable, the results will be bias Thus, Random effect probit model is utilized The result is presented with random effects integrated into probit models and takes for granted that random effect exists and correlated with other determinants

In the study, author mentioned two problem to be tested The first is how difference is the model if univariate probit is applied for each aspect of innovation instead

of a pooled probit The second question is the interaction between managerial skill and financial resources The result proves that correlation between unobserved factors and between two options of innovation doesn't exist

The analysis of two innovation activities allows author to study interactive

relationship between product and process innovation However, innovation dummy

variable takes two value only, so it doesn’t reflect the nature of innovation is improving existing products or creating new ones

This part presents a review of empirical studies about determinants of product innovations This is an important references for establishing an analytical model Many factors are suggested to have influence on innovation activity These factors are divided

to characteristics of firm or of entrepreneur such as firm size, characteristics of owner/ manager, human capital, investment on training and marketing, origin of ownership, R&D activities The second group relates to characteristics of market or information sources outside such as competition level, outsourcing and employing external assistant, networking Among all determinants, some are expected to have positive impact: technical staff, managerial and professional staff, training cost, marketing cost, export orientation, R&D expenditure, level of competition Some are proved to have negative influence: outsourcing level, age of entrepreneur The effects of some factors are even ambiguous such as firm size There are sufficient empirical review for these variables to

be examined in the model of this paper However, feasibility of each factor also depend

on availability of data This issue will be clarify in next part of this paper

Figure 1: Conceptual Framework

Product innovation:

* Improving existing products

* Creating new products

External factors

Outside sources of information

* Outsourcing

* Networking

* Technical consultant

* Knowledge of market/ Marketing cost

* Qualified technical staff

* Managerial and professional staff

CHAPTER III: RESEARCH METHODOLOGY AND DATA

III.1 Data and sample

This study utilizes data collected via SME survey in Vietnam in 2011 for 2552 firms The survey is carried out by Vietnamese Institute for Labour and Social Affairs (ILSSA), Ministry of Labor, Invalids and Social Affair (MOLISA) in corporation with University of Copenhagen, Denmark Data was taken from official website of University

of Copenhagen http://www.econ.ku.dk/derg/link/Vietnam/ The survey gather answer from 2552 participants who is manager/owner of SME firms across ten provinces of Vietnam which includes Ha Noi, Hai Phong, Ha Tay, Phu Tho, Quang Nam, Nghe An, Khanh Hoa, Lan Dong, Ho Chi Minh City and Long An The total survey includes 2552 observations However, only 1418 observations have no missing value Thus, these 1418 observations are used as sample for the regression

This study focus on four major industries: food processing and producing, Textile

& shoes manufacturing, mechanic industry, and wood industry The reason is these four sections are among major industries of Vietnam which account for 70% of total export value of Vietnamese firms In general, there are 1418 observations, in which 302 from food section, 102 from textile section, 240 from mechanic section, 133 from wood section and 641 from other section

III.2 Variables and Measurement

Below is all variables including independent and dependants variables that will be added into the regression Independent are categorized into two groups: Internal factors relating to firm's characteristics and external factors involving in market's features or outside sources of information

III.2.1 Dependent variable

Product innovation may take two options: improving current product or making new product Thus, there are two dependent variables Each option is measured by a dummy variables which have two discrete value 0 or 1 When product is improved or