The influence of country education level on the relationship
between innovative activities and firm performance
An investigation of Central and Eastern European
food-manufacturing companies
By Yolanda Hut S2165325 y.hut@student.rug.nl
Thesis MSc International Business and Management University of Groningen
Faculty of Economics and Business
13th of June 2016
Word count: 13399
The influence of country education level on the relationship
between innovative activities and firm performance
An investigation of Central and Eastern European
food-manufacturing companies
Abstract: Human capital within firms positively influences the outcomes of innovative activities. This study extends these findings by investigating whether the average education level of countries influence the effect of innovation activities on increased sales performance. This study is executed by using a dataset from the BEEBS survey over the period 2012-2014; including 179 food-manufacturing companies within Central and Eastern Europe. In this region several changes have advanced the food-industry towards a more dynamic industry where innovative activities become highly required. To measure innovative activities, four categories of innovation (product, process, marketing and organisational) are merged into one variable. Moreover, it is tested whether food-manufacturing firms should focus on a certain combination of the four innovative activities to generate the highest sale increase. The results show that higher education indeed significantly stimulates the effect of innovation activities on sales performance. Moreover, for food manufacturing it turns out that the combination of product, process and marketing innovation leads to the highest increase in sales performance.
TABLE OF CONTENTS
LIST OF TABLES ... 5 LIST OF FIGURES ... 5 1. INTRODUCTION ... 6 2. LITERATURE REVIEW ... 9 Food-manufacturing industry ... 9Food-manufacturing industry in Central and Eastern Europe ... 10
Resource based view and dynamic capabilities ... 10
Innovative activities ... 12
Types of innovation ... 12
Innovation in the food-manufacturing industry ... 14
Measuring innovation performance within the food-manufacturing industry ... 15
Knowledge and innovation ... 16
Conceptual model ... 17
3. METHODOLOGY ... 19
Research design ... 19
Research philosophy and approaches ... 19
Research strategy, choices and time horizon ... 19
Data collection ... 20
Variables hypothesis 1 ... 20
Variables hypothesis 2 ... 21
Control variables ... 21
Data analysis ... 23
Reliability and validity ... 25
4. DATA ... 26
Descriptive statistics ... 27
Firm characteristics ... 28
Main results ... 30 Hypothesis 1 ... 30 Hypothesis 2 ... 30 Discussion of findings ... 33 6. CONCLUSIONS ... 35 Theoretical implications ... 35 Managerial implications ... 36
Limitations and avenues for further research ... 37
REFERENCES ... 39
APPENDIX A – CORRELATIONS ... 45
APPENDIX B – EXCHANGE RATES ... 46
LIST OF TABLES
Table 1 – Explanation of the four types of innovation ... 14 Table 2 – Combination of innovation activities that need to be tested ... 22 Table 3 – Operational definitions of dependent, independent, moderator and control variables ... 22 Table 4 – Descriptive statistics of dependent, independent, moderator and control variables ... 27 Table 5 – Company spread among countries and average education levels ... 28 Table 6 – Division of (international) markets for the main products of the companies ... 29 Table 7 – Moderator analysis on overall innovation activity x macro education level ... 32 Table 8 – Results of One Way ANOVA tests: effect of types of innovation on sales performance ... 32
LIST OF FIGURES
1. INTRODUCTION
With a turnover of €1,062 billion and 288,655 enterprises in 2015, the food-manufacturing industry is one of the most important industries regarding output and employment within Europe (ECSIP consortium, 2016). However, even though the food industry within Europe is tremendous and is able to introduce a lot of new products, the industry used to be characterized by low R&D intensity and the introduction of radical innovations was low (Galizzi & Venturini, 2012). Nonetheless, recent technical and economic changes in society and in the industry make the industry more and more technology intensive (Bigliardi & Galati, 2013; Traill & Meulenberg, 2002).
Changes such as access to the single market within Europe, standardization of policies and privatizations have greatly impacted the food-manufacturing companies in Eastern and Central European countries (Dries, Reardon & Swinnen 2004; Traill & Pitts, 1998). Accordingly, increased competitiveness and enhanced international opportunities raise the pressure to have resources and capabilities that make the companies able to innovate more quickly and efficiently to build a sustainable competitive advantage (Traill & Pitts, 1998).
According to the resource-based view, the accumulation of valuable, rare, inimitable, non-substitutable (VRIN) and well-organized resources is enough to build a sustainable competitive advantage (Barney, 1991; Barney, 1995). However, as the food-industry is becoming increasingly dynamic, a sole focus on the accumulation of this type of resources is not enough anymore (Eisenhardt & Martin, 2000). The companies have to develop dynamic capabilities, since these will help them create sustainable competitive advantages based on three specific dynamics: processes, positions and paths (Teece, Pisano & Shuen, 1997). Possessing these capabilities should help the food-manufacturing firms to be timely responsive and be able to have rapid and flexible innovations (Teece & Pisano, 1994).
Especially, for the food-industry there is a tension between two types of innovation that require special innovative activities. Food manufacturers face a trade-off between technology-push innovation and demand-pull innovation (Grunert & Traill, 2012). The first one involves the adaption towards technological change, which makes the firm able to develop new products or improve existing products and processes based on new technological knowledge. Besides, demand-pull innovation considers innovation as a detection of the fulfilment of unfulfilled needs of new (potential) customers. For this type of innovation, new product development is not enough, but also new product management and thus marketing and organisational innovation are needed (Grunert & Trail, 2012). When firms are able to execute all these innovative activities successfully, it should be possible to measure an increase in firm performance and hence in sales performance (Evangelista & Vezzani, 2010).
However, to achieve this increase in firm and sales performance, dynamic capabilities are the key. An important way to obtain dynamic capabilities is to increase the knowledge base of the firm (Teece & Pisano, 1994). Related to this extra knowledge-gain that is required, absorptive capacity is an essential factor. Absorptive capacity is the ability of a firm to recognize the value of new external information, assimilate it and apply it to commercial ends (Cohen & Levinthal, 1990). Important for the development of absorptive capacity within firms is prior related knowledge (Cohen & Levinthal, 1990). Lund Vinding (2006) found that in the service and manufacturing industries the human capital within firms, which included among others the share of highly educated employees, positively influences the absorptive capacity of firms. This increased absorptive capacity in turn resulted in an increased ability to innovate. Even though Lund Vinding measured education within the firm, and other studies also researched human capital within firms, no study has made a comparison across countries and investigated whether the average education level of the country is able to stimulate successful innovative activities of firms. Therefore, the goal of this study is to investigate the relation between innovation activities and sales performance moderated by the country education level within the food manufacturing industry. Hence, the research question is:
Do the innovation activities of companies in the food manufacturing industry generate higher sales performance when influenced by a higher country education level?
types of innovation activities are found to be important within the food-manufacturing industry, this study will additionally investigate which type or combination of innovation activities leads to the highest increase in sales performance.
This study aims to contribute to the existing literature in several ways. First of all, this study uses a new combined measurement for innovative activities, namely the combination of product, process, marketing and organisational innovation. There has not been an empirical study before that investigated the combinations of those four specific innovation types in relation with sales performance. Further, this study shows how important it is to segment innovation activities into distinctive categories as it leads to big differences in increased sales performance. Moreover, this study shows the implications of high education levels on increased innovative activities and sales performance of firms. Studies on human capital and innovation performance are executed before. However, using average country education level as a moderating variable is new, which extends literature on strategic innovation management.
From a managerial perspective, managers from food-manufacturing companies and policy makers in Central and Eastern European countries learn how important it is to have a high overall educated workforce. It will help managers and policy makers to set the best business environment related to advancing education levels within their countries. When firms and policy makers understand the importance of higher education levels this may encourage them to recruit and lobby for higher educated workforces, which will eventually be beneficial for the firms and the whole country. Moreover, the food-manufacturing firms will understand that it is not enough to focus on just one innovative activity, but that the combination of product, process and marketing innovation will work best to enhance the sales performance of the firm.
2. LITERATURE REVIEW
The main objective of this study is to investigate the dynamics between the innovative activities of food-manufacturing firms and their influence on sales improvements influenced by the education level of the country. Accordingly, this chapter will take a broader look into the theoretical backgrounds of these phenomena.
Food-manufacturing industry
Within Europe the food industry is the largest manufacturing sector in terms of output value (Traill & Pitts, 1998), and of one the most important branches regarding employment. In the year 2012, the food and drink manufacturing industry realized a turnover of 1,062 billion euro, with a total of 288,655 enterprises and 4,530,000 employees working in this sector. Moreover, the turnover grew by almost 7 percent in the years 2008 till 2012 (ECSIP consortium, 2016).
Even though the food industry within Europe is tremendous and is able to introduce a lot of new products, the industry was characterized by low R&D intensity and the introduction of radical innovations was rather low (Galizzi & Venturini, 2012). Therefore, the industry was still regarded as a low-tech industry; however, the industry is becoming more and more technology intensive (Traill & Meulenberg, 2002). Technical and economic changes in society and manufacturing have influenced the entire food supply chain (Bigliardi & Galati, 2013). Nowadays, for the food-manufacturing industry, product development and innovation are seen as essential factors to survive in the competitive world (Grunert & Traill, 2012). Several factors such as using new ingredients and improvements in the processing and packaging procedures have accelerated the new product development within the food manufacturing industry (Galizzi & Venturini, 2012).
opportunities within the food-manufacturing industry (Traill & Pitts, 1998). All these major changes encouraged the shift to more technology intensive manufacturing with increased competiveness and international opportunities (Traill & Meulenberg, 2002).
Food-manufacturing industry in Central and Eastern Europe
Countries that more recently got access to the European Union, its single market and its agricultural policies, such as Central and Eastern European countries, face the increase in competitiveness even more because of the new international opportunities and standardization of rules and policies within Europe (Traill & Pitts, 1998). Besides, these countries faced the privatization of many sectors such as the retail industry and in particular the supermarket sector (Dries et al., 2004). The rapid rise of the supermarkets gave the retail sector more power and made it possible to select preferred producers based on high quality and safety requirements (Dries et al., 2004). To meet all these standards and to face the increasing pressure of competitiveness, food-manufacturing companies within Central and Eastern Europe face a major pressure to have specific resources and capabilities that make them able to innovate more quickly and efficiently and thus to sustain their competitiveness in this industry (Traill & Pitts, 1998).
Resource based view and dynamic capabilities
The resources and capabilities that are needed for those innovations and to increase competitiveness are based on all the financial, physical, human and organizational assets used to develop, manufacture and deliver the products to the customers (Barney, 1995). According to the resource-based view, the accumulation of valuable, rare, inimitable, non-substitutable and well-organized resources is enough to build a sustainable competitive advantage (Barney, 1991; Barney, 1995) and to increase the overall firm performance (Lin & Wu, 2014).
The resource-based view works well in low-velocity markets where it is possible to leverage on the identified strategic dynamics related to valuable, rare, inimitable and non-substitutable resources and competences (Eisenhardt & Martin, 2000). However, in dynamic markets it is not possible to leverage solely on these resources. The fast-changing environments require firms to have capabilities that make it possible to build, integrate and reconfigure resources at each point in time. This is needed to respond quickly to changing dynamics within the markets (Eisenhardt & Martin 2000; Lin & Wu, 2014).
marginal tensions to increase competitiveness. However, recent changes have made this industry more dynamic and vulnerable for environmental changes with increasing competitiveness (Galizzi & Venturini, 2012; Traill & Pitts, 1998; Traill & Meulenberg, 2002). Therefore, solely focusing on the accumulation of resources, as identified by the resource-based view, is not enough anymore to survive as a firm within the food-manufacturing industry. As argued by Teece et al. (1997), a sole focus on the accumulation of valuable (technology) assets is not sufficient to become a sustainable competitive firm.
Companies should work on developing dynamic capabilities, since these will help them create sustainable competitive advantages (Teece et al., 1997). The new winners can be identified by firms that are timely responsive and able to have rapid and flexible product innovations. Moreover, they possess the management capability to effectively coordinate and redeploy internal and external competences (Teece et al., 1997). These capabilities rest in the three dimensions described by the dynamic capability view: processes, positions and paths (Teece & Pisano, 19954). First, the processes involve the routines of the firm and how things are done. Secondly, the position includes the firm’s assets, its endowments of technology, all the possessed intellectual property, its customer base and also the external relations with suppliers and other parties. Lastly, the paths include the strategic alternatives available to the firm (Teece & Pisano, 1994; Teece et al., 1997). Together these three capabilities constitute the dynamic capabilities. It makes firms able to integrate, build and reconfigure internal and external competences to address rapidly changing environments and thus to achieve innovative forms of competitive advantages (Teece et al., 1997). Within dynamic environments, the development of dynamic capabilities, and thus also the on-going execution of innovative activities is seen as crucial in order to survive in the dynamic business environment (Subramanian & Nilakanta, 1996; Teece & Pisano, 1994). As the food-manufacturing industry has entered a more dynamic environment, a focus on the development of dynamic capabilities and innovative activities is becoming increasingly important.
ends (Cohen & Levinthal, 1990). Absorptive capacity is seen as one of the most important dynamic capabilities pertaining to knowledge creation and utilization (Zahra & George, 2002). It is critical for the innovative activities of the firm and can be seen as one of the determinants why firms are more successful than others in creating competitive advantages in dynamic markets (Cohen & Levinthal, 1990; Lin & Wu, 2014).
These arguments that support the development of dynamic capabilities do not imply that the focus on valuable, rare, inimitable and non-substitutable resources can simply be ignored. The dynamic capabilities can mediate these resources to improve firm performance (Lin & Wu, 2014). Especially learning and knowledge are major dynamic capabilities that mediate a positive influence of VRIN resources on firm performance (Lin & Wu, 2014).
To conclude, dynamic capabilities can help to enhance and reconfigure resources and execute innovative activities, leading to increased firm performance (Eisenhardt & Martin, 2000; Lin & Wu, 2014). Since, the food-manufacturing industry nowadays faces the pressure to renew and stay competitive, dynamic capabilities and thus learning and knowledge accumulation becomes increasingly important. These capabilities will help to execute innovative activities, which are needed to survive in the food-manufacturing industry.
Innovative activities
Innovation involves the implementation of an idea into a new device or process for applications in commercial or practical objectives (Schilling, 2013) with the purpose to create new business (Trott, 2008). It is seen as one of the most important drivers for competitive success (Schilling, 2013), which also yields for the food-manufacturing industry (Grunert & Traill, 2012).
Types of innovation
The innovative activities of a firm are often divided into two different activities: product innovation and process innovation (Schilling, 2013). Product innovation refers to the direct output of organizations, namely the newly introduced goods or services. While process innovation refers to improving the effectiveness or efficiency of the production process. Both are seen as extremely important for a firm to compete and they often occur simultaneously (Schilling, 2013).
increase new product performance, several factors are found to be important: having a high-quality new product process, a clearly defined new product strategy and also having a devotion of enough resources related to money and human capital. Moreover, research has shown that for successful new product development, multidisciplinary cooperation between departments is necessary (Olson, Walker, Ruekert & Bonner, 2001). Therefore, during the whole process of new product development, in-depth cooperation between marketing, operations and R&D is needed to reach the optimal result. Especially the harmonisation between marketing and R&D is found to be of great importance for new product innovations (Souder, 1988).
Besides the distinction of product and process innovation, Evangelista and Vezzani (2010) stress a distinction between technological innovations and non-technological innovations. According to them, product and process innovation can be classified as technological innovations. The technological novelty based from product innovation should lead to a competitive advantage, while the efficiency and productivity gains should be the main driver for creating a competitive advantage within process innovation. Together these drivers should lead to extraordinary results in growth of sales and market share gains, but not without non-technological innovations (Evangelista & Vezzani, 2010). Non-technological innovations such as organisational and marketing innovations can be seen as complements of technological innovations. Organisational or management innovation involves the improvements in new ways of working in the top level, possibly leading to product and process improvements (Laforet, 2011). Marketing innovation is seen as an incremental innovation type that involves the improvements in product design, placement, promotion or pricing (Naidoo, 2010). Accordingly, marketing innovation can help improving sustained competitive advantage since it helps delivering better (perceived) value to the customer, without necessarily big improvements in the product (Naidoo, 2010). The two additions of marketing and organisational innovations amplify the findings from Olson et al. (2001) and Souder (1988) that multidisciplinary cooperation between R&D, operations and marketing are necessary for successful innovation.
Innovation in the food-manufacturing industry
Grunert and Traill (2012) argue that for the food industry innovative activities are essential factors to survive in the global competitive world. What makes this industry interesting is the fact that changes within the last decades have made the industry more R&D intensive and competitive. Based on stronger requirements from retailers and consumers regarding product and process improvements, companies are obliged to execute innovative activities (Dries et al., 2004; Traill & Pitts, 1998).
Likewise, for the food-manufacturing industry the innovation process does not only involve the development of new products. It increasingly involves the right and new targeting tactics to bring the products to new customers (Traill & Meulenberg, 2002). Accordingly, solely product and process innovations are not enough to survive in the competitive world within the food-manufacturing industry. Grunert and Traill (2012) describe that food manufacturers face a trade-off between technology-push innovation and demand-pull innovation. The first one, technology-push innovation, involves the adaption towards technological change, which makes it possible to develop new products or improve existing products based on new technological knowledge and processes. On the other hand, demand-pull innovation considers innovation as a detection of the fulfilment of unfulfilled needs of new (potential) customers. For this latter one, new product development is not enough, but also improved product management is needed (Grunert & Traill, 2012), which relates to organisational and marketing innovation. Therefore, sole R&D activities related to new and product-releases and improved processes are not enough. Also marketing and organisational skills need to be developed and improved (Traill & Meulenberg, 2002).
Therefore, food manufacturers should also get involved in innovative activities related to marketing and organisational aspects as introduced by Naidoo (2010) and Laforet (2011). Accordingly, it can be expected that the innovation activities of food-manufacturing
Type of innovation Definition
Product innovation Relates to the direct output of organisations, such as newly introduced goods or services
Process innovation Relates to improving the effectiveness or efficiency of the production process
Organisational innovation Relates to the improvements in new ways of working, possibly leading to product and process improvements
Marketing innovation Involves the improvements in product design, placement, promotion or pricing
industries should include (the package of) product innovation, process innovation, marketing innovation and organisational innovation in order to have the biggest improvement in firm performance. However, as innovation may take a lot of resources and time (Schilling, 2013), it can be expected that not all firms are able to execute all those activities simultaneously. Nonetheless, there is no consensus on which innovation activity or which combination of innovation activities is the most important activity for increased firm performance within the food-manufacturing industry yet.
Measuring innovation performance within the food-manufacturing industry
As innovation leads to new businesses and ideas, it can be generally expected that it should lead to increased firm performance. However, measuring firm performance based on innovative activities or new product introductions may be difficult. Cordero (1990) argues that one of the most important aspects in measuring innovation performance is using a systematic measurement model. Evaluating a new product development process is mostly based on measuring the effectiveness and efficiency of the whole process, while measuring the overall innovation performance is mostly based on return-related dimensions (Schilling, 2013). Measurements for overall innovation performance can be the return on innovation, the percentage of projects that achieve their sales goals, which percentage of revenues are generated by products developed within the past years, or the success of the new projects in relation to the total product portfolio (Schilling, 2013).Moreover, Adams, Bessant and Phelps (2006) identified a framework with seven categories to measure innovation management: inputs management, knowledge management, innovation strategy, organizational culture and structure, portfolio management, project management and commercialization.
commercialization and thus an increase in sales will be an appropriate measurement to evaluate innovation performance within the food-manufacturing industry.
Knowledge and innovation
To achieve the positive results with the innovation activities, one of the most important dynamic capabilities that is needed is absorptive capacity (Zahre & George, 2002). Zahra and George (2002) have found significant results between absorptive capacity and successful innovative output. Moreover, the combination of absorptive capacity with the possession of VRIN resources should help in creating competitive advantages based on improved innovative capacities (Cohen & Levinthal, 1990; Lin & Wu 2014).
To create the highest possible absorptive capacity, a broad level of prior related knowledge and background diversity of employees are needed (Cohen & Levinthal, 1990). The ability to learn, assimilate and apply new knowledge leads to an increase in overall intellectual capital of the firm. Intellectual capital can be seen as one of the major forces that drive the firm performance and earnings; it guides the firm towards the future (Sullivan, 1998). Sullivan (1998) defines intellectual capital as the knowledge of the firm that can be converted into profits. Intellectual capital can be divided into two parts: the human capital and intellectual assets. First, the human capital consists of the individual overall knowledge of employees, related to its skills, abilities, experiences and know-how. Second, the intellectual assets are based on the human knowledge that is shared, made available and usable for the firm. To develop human capital, experience and education are essential factors (Becker, 1962). People with higher education levels are assumed to have more extensive experiences and invest more time, energy and resources in skills development, resulting in higher contributions to the firm (Dakhli & De Clercq, 2004). Accordingly, when there are more highly educated people in the firm, the stock of internal knowledge will increase and also more network opportunities to external knowledge will be facilitated; resulting in a higher level of absorptive capacity of the firm (Mangematin & Nesta, 1999).
individuals are needed for R&D activities (Vega Jurado, Gutiérrez Gracia & Fernández de Lucio, 2008). However, important to note is that Thornhill (2006) found that this positive relationship between high human capital and innovation only yields for dynamic environments. High knowledge assets of a firm in relation to innovation seem to be of greater importance in dynamic environments than in stable environments. For stable environments the benefits regarding firm performance and innovation are higher when focusing on training in stead of on knowledge assets (Thornhill, 2006).
Since the food-manufacturing industry in Central and Eastern Europe is shifting towards more intensive technologies and dynamic markets, these results might indicate the increasing importance of a high education level for the development of dynamic capabilities and accordingly innovative activities within the food-manufacturing industry.
Conceptual model
Based on all discussed literature several important factors for firm performance in relation with innovative activities can be distinguished. First, the literature has identified that the food-manufacturing industry in Central and Eastern Europe faces pressures to develop dynamic capabilities needed for efficient and quick innovations. Moreover, the food-manufacturing industry experiences tensions that result in the need for distinctive innovative activities: product, process, marketing and organisational innovations are all essential. To be able to execute these innovative activities in the best way it is important to develop dynamic capabilities that help in applying and reconfiguring resources in the best possible way. However, to be able to develop those dynamic capabilities, it is extremely important to have adequate human capital, which is partly based on the education level of the employees. As the food-manufacturing industry moves towards a more dynamic environment, the relation between human capital and innovation becomes even of greater importance. Therefore, it is interesting to see whether overall education levels of countries impact the relation between innovation activities and sales performance within the food-manufacturing industry.
The model suggests that for firms the overall country education level may be an important moderator to increase the firm (sales) performance based on the executed innovative activities. According to the literature, high human capital within the company positively impacts absorptive capacity, R&D activities and innovative performance. However, those studies always focused on the knowledge inside the firm, without considering the macro conditions of the country. Therefore, this study takes a different perspective and investigates the overall education level per country as a moderator of the relation between innovation activities and sales performance. Moreover, within this study the innovative activities are measured in four distinctive activities: product, process, marketing and organisational innovation. As is argued in the literature, all the four innovative activities are highly recommended to execute within the food-manufacturing industry. Therefore, it can be argued that when a firm is able to focus on all the four innovative activities simultaneously, the return on sales should be the highest. Until now, no research on which innovative activity or combination of activities may have the highest result on sales performance of the firm exists. Especially for food-manufacturing firms with Eastern and Central Europe, who face rising competitions and high demands to invest in new technologies and innovations, it may be interesting to investigate whether it is important to focus on all four innovative activities, or that another combination of innovation activities will be better.
Therefore, the following two hypotheses are stated:
H1: Innovation activities generate higher sales when the education level in the country is higher.
H2: When executing all the four innovation activities simultaneously, firms within the food-manufacturing industry are able to generate the highest sales increase.
Figure 1 – Conceptual model
3. METHODOLOGY
The purpose of this study is to get insight in the sales performance of firms based on their innovative activities conditioned by the education level of the country. Based on this purpose, two hypotheses are developed, which test whether education level moderates the impact of innovative activities on sales performance (h1) and whether executing all four innovative activities simultaneously leads to the highest increase in sales performance (h2).
Based on the investigated literature, the developed conceptual model and the two hypotheses a solid methodology has been chosen. This section discusses the chosen research design, the data collection, the analyses procedures of the hypotheses and lastly the reliability and validity of the estimation methods.
Research design
Research philosophy and approaches
The methodology of this study is based on the view of positivism and the empirical cycle, where factual knowledge about the real world will be gained by the use of survey data (van Aken, Berends & Van der Bij, 2012). Moreover, a deductive and quantitative approach is used. Based on the literature, two hypotheses are developed. These hypotheses are empirically tested by using surveys and country-level data. The results of these tests are evaluated and interpreted by comparing and relating it to existing literature.
Research strategy, choices and time horizon
The survey data of this study is based on two databases provided by the European Bank for Reconstruction and Development and the UNESCO Institute for Statistics (UIS). Based on those two databases 179 food-manufacturing companies within Central and Eastern Europe are investigated. By using the survey data of food-manufacturing companies within Central and Eastern Europe and the IUS data, causative relations between innovation, sales performance and education levels can be examined. Since this study only contains statistical data, a mono method is used.
Data collection
The unit of analysis in this research are the food-manufacturing companies within Central and Eastern European countries. Based on the unit of analysis, the important variables have been selected and the datasets have been made ready for the analysis.
All the variables are based on questions within the BEEBS survey and IUS database and therefore, their operational definitions are used. Hence, to test the outlined hypotheses all the variables within the conceptual model have been operationalized. The operational definitions and accompanying answer categories or outcome ranges can be found in table 3 on page 22.
Variables hypothesis 1
The first hypothesis tests whether innovation activities generate higher sales when a higher education level in the country is present. Based on this assumption it can be derived that the independent variable concerns innovation activities of the firm, the moderator variable is education level and the dependent variable is sales performance.
First, to measure innovation activity a combined variable of all four innovation types is used. As the literature has identified, all four innovative activities (product, process, marketing and organisational) are extremely important for companies within the food-manufacturing industry. Therefore, it is an appropriate measurement to merge these four variables into one overarching innovation measurement. Hence, a new variable that consists of the combination of the four innovation types is built. The combination of those four identified innovations types has a Cronbach’s alpha of 0.764. Therefore it can be concluded that this new variable is a reliable measurement for overall innovation activities.
Secondly, the moderator variable involves the education level per country of the involved companies. This macro indicator displays the percentage of population aged over 25 years that completed a tertiary education. Tertiary education is defined as “education that builds on secondary education, providing learning activities in specialised fields of education. It aims at learning at a high level of complexity and specialisation. It includes as what is commonly understood as academic education but also includes advanced vocational or professional education (UIS, 2011).”
are asked. First, it is asked what the number of sales for last year was, followed by a question concerning the number of sales three years ago. By subtracting the sales in year t-3 from year
t, a new measurement is created. During the study all the sales amounts are measured in local
currencies. However, for the purpose of this study all amounts are converted into Euros. How this is executed is explained further in section 4 on page 26.
Variables hypothesis 2
The second hypothesis investigates which combination of innovation activities generates the highest sales improvement and what the significant effect is. This study concerns four types of innovation as independent variables: product, process, marketing and organisational. All the four variables are dichotomous measurements. When the variable has a value of zero it means that the firm did not execute this type of innovation. Accordingly, a value of one means that the firm did execute this type of innovation. For this hypothesis, all the possible combinations of innovative activities are tested. The hypothesis proposes that the combination of all the four innovation activities should lead to the highest increase in sales. However, to be able to test this and to get a comprehensive understanding of the sales increase based on innovative activities, all the combinations of possible innovation activities are tested. The combinations of variables that are tested are presented in table 2 on page 22.
Control variables
Data analysis
To tests the two hypotheses, two different analysis methods are used. This section will shortly discuss the two estimation models.
Since the first hypothesis concerns the effect of an independent variable (innovative activities) on a dependent variable (sales performance) moderated by another variable (education level), a moderator analysis is used. The relationships within a moderator analysis are presented in figure 2 below:
T
This figures shows that the relationship between two variables, innovative activities and sales performance, depends on a moderator variable, in this case the education level. For this moderator variable, a new interaction predictor is calculated. This interaction predictor is based on the interaction between the independent variable (innovative activities) and the moderator variable (education level). To calculate this moderating effect, the most appropriate way will be using linear multiple regression (Cohen, Cohen, West & Aiken, 2013). The statistical formula for this calculation is presented as formula 1:
! = !0 + !1!1 + !2!2 + !3 (!1 ∗ !2) + ∈ (1) In this formula, Y represents the dependent variable sales performance, β0 is the intercept, β1 represents the coefficient of the overall innovation activity, β2 represents the coefficient of education level and β3 the coefficient of the interaction between overall innovation activity and education level. Moreover, x1 and x2 represent respectively the explanatory variables of overall innovation activity and education level. Lastly, ∈ expresses the standard error of the formula. Since this analysis includes a discontinuous independent variable, and a continuous moderator variable, both variables are standardized to come to a correct interaction variable that represents both variables in an equal way (Cohen et al., 2013).
Innovative activities Education level Innovative activities x Education level Outcome variable
The second hypothesis concerns a comparison of means (sales performance) between two different groups, namely the group companies that do execute certain innovation activities, and the group of companies that do not execute those activities. Comparing the sales performances of these groups gives a good indication whether innovative activities help to increase sales performance.
Since the goal of this hypothesis is to investigate which innovative activities generate the highest sales performance, a comparison of all those means is made. First, the individual innovative activities (product, process, organisation and marketing) and their impact on sales performance are measured. This is executed by calculating one-way ANOVAs. One-way ANOVAs are based on formula 2:
!!" = ! + !!+ !!" (2)
This linear model represents xij, which represents the individual observations, as the
sum of three parts: (1) µ represents the overall mean of the observations, (2) aj stands for the
treatment effect and eij (3) represents the random element for a normally distributed
population (Armstrong, Eperjesi & Gilmartin, 2002). Based on all those observations, a F-statistic can be calculated. This F-F-statistic is based on the calculation of the mean squares of the treatment effect and of the error effect. This F-statistics shows whether the results of the means differ significantly or not. Therefore, based on the calculated means and the F-statistic it is possible to interpret whether the independent variable makes a significant impact on the dependent variable.
When combining two or more innovative activities, a one-way ANOVA is not sufficient anymore and a two-way ANOVA is used. Two-way ANOVAs control for the interaction effects between the various independent variables and check whether these interaction effects also significantly impact the dependent variable. Therefore, for each combination of innovative activities, a two-way ANOVA is calculated.
The calculation of a two-way ANOVA is based on formula 3:
!!"#= ! + !!+ !! + (!")!"+!!" (3)
This linear model is essentially the same as formula 2, however, now a second variable is added (!!) and also the interaction effect between these two variables is added (!")!". If more variables are added, the formula will be extended with more independent and
however, for this research only the statistic for the interaction effects is important. This F-statistic is calculated by dividing the mean square of the interaction effect by the mean square of the error effect. After the two-way ANOVAs for all the possible combinations have been executed, the means and significance of all these tests are compared to discover which innovative activities, or combinations of innovative activities, contribute mostly to sales performance of firms within the food-manufacturing industry in Central and Eastern Europe. Reliability and validity
To be confident that the results are reliable and valid, several pre-tests are conducted. Especially, for the first hypothesis, which concerns a regression analysis, it is important to conduct several pre-tests to understand how reliable the results are. Since the independent variable in this hypothesis concerns a new variable that is made up of four variables, a Cronbach’s alpha has been conducted. The Cronbach’s alpha tests for internal consistency and validity of the new developed variable (Blumberg, Cooper & Schindler, 2005). The Cronbach’s alpha of the new variable is 0.764, and thus it can be concluded that the new variable is a good measurement to represent the four innovation activities.
Next, for a regression analysis it is important to check for multicollinearity. Multicollinearity arises when inter-correlations between the two predictors (independent and moderator variable) are very high (Malhotra, 2010). If this is the case, it may become difficult to estimate the regression coefficient precisely, as the standard errors are likely to be high. Moreover, when multicollinearity arises the magnitudes and signs of the regression coefficients may change for each sample, which make it difficult to assess the importance and validity of the independent variables in explaining the variation in the dependent variable (Malhotra, 2010). First it is tested whether a correlation between the independent and moderator variable exists. This correlation table can be found in appendix A. The Pearson correlation between the education level and innovation variable is 0.025, however, not significant. Secondly, the multicollinearity test for the regression in this study gives a VIF value of 1.001, which means we can assume that multicollinearity is not a problem.
4. DATA
As has been shortly described in the previous chapter, the data for this study is based on two databases: the BEEPS database of the European Bank for Reconstruction and Development and the UNESCO database. The BEEPS database provides data for the years 2012-2014. The European Bank for Reconstruction and Development and the World Bank Group initiated this survey to get a better understanding of firm’s perceptions of their operating environment. Ultimately it should help policy makers to better understand how businesses experience the business environment; in this case related to the innovative activities resulting in better firm performance. Moreover, the data concerning the education level has been gathered by a database made available by the United Nations Educational, Scientific and Cultural Organization (UNESCO) Institute for Statistics. It includes educational attainment percentages of 149 countries (UIS, 2016). For this research the education level is based on the percentage of the population in each country that completed a tertiary education in the year 2012 (Croatia 2011 and Belarus 2009, since they have no data available of 2012). The data regarding education levels can be found in table 5 on page 28.
It has appeared to be difficult to find complete agreement on which countries exactly are classified as Central and Eastern European countries. For this study the combined definitions according to the Organisation for Economic Co-operation and Development (OECD) and the United Nations Statistics Divisions have been used. This has lead to the following countries that will be analysed: Albania, Belarus, Bulgaria, Croatia, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Republic of Moldova, Romania, and the Slovak Republic (excluding Russia and Ukraine; no suitable data available) (OECD, 2001; UN Statistics Division, 2013).
The numerical values in the BEEPS database are based on the local currency for each measurement. Therefore, the data regarding sales performance has been adjusted to the Euro. The exchange rate for these adjustments is based on the average exchange rate for each country in 2013 since almost all measurements took place in 2013. The used exchange rates can be found in appendix B. Moreover, to make the tables and information regarding sales performance more readable; all the sales performance values are divided by 1,000.
Descriptive statistics
To create a short overview of all the descriptive statistics for this study, table 4 below provides a summary with all the relevant data for the used variables in this study.
Variable Observations Mean St. Dev. Min Max
Product innovation 177 0.40 0.492 0 1 Process innovation 178 0.30 0.459 0 1 Organisational innovation 177 0.21 0.412 0 1 Marketing innovation 176 0.28 0.452 0 1 Education level 179 0.22 0.092 0.1290 0.5166 Sales performance (x1000) 179 681.29 3,322.62 -13,418.01 24,055.96 Company size 179 1.73 0.818 0 5
The table presents the number of observations, the means, standard deviations and the minimum and maximum value of each variable. As the variables regarding innovation are dichotomous they only continue values between 0 and 1. Since product innovation has the highest mean, it can be concluded that product innovation has been executed most often among the companies, and organisational innovation least often. Moreover, it can be seen that the sales increase of the firms is on average €681,290. However, considering the large standard deviation, it is important to interpret this mean with caution.
Moreover, a correlation analysis has been executed to test whether correlation between the independent and dependent variables exist. The results of this test can be found in appendix A. The results show that for the variables of the regression analysis, the overall innovation activity and the education level, the correlation variable is low (0.025). Moreover, the correlations between the individual independent variables and dependent variable appeared to be significant but not really high (0.256 for education level and sales performance, and 0.209 for overall innovation activity and sales performance).
As this study involves a comparison among countries in Central and Eastern Europe, it is helpful to know how many companies are analysed per country. Therefore, table 5 on page 28 shows the division of the companies among the thirteen countries. It can be observed that most companies of this study reside in Moldova (30), and least in Bulgaria (7).
percentage of 51.66 percent of the citizens that completed a tertiary education. In Albania only 12.90 percent of the citizens completed a tertiary education.
Firm characteristics
Before this study can proceed to the analysis regarding innovation and firm performance, it is useful to have a bit more detailed information about the companies within this study.
To start with, it can be observed that most companies in this study (41.9 per cent) consist of small companies with 5 till 19 employees. Next to that, 34.6 percent of the firms are medium-sized with 20-99 employees and 20.7 per cent of the firms have over 100 employees. Only 2.8 percent of the companies have 4 or fewer employees. As there is quite some spread between the companies, and the company size may highly influence sales performance of firms, the company size will be included in the analyses and used as a control variable.
As has been presented in the literature, pressures to innovate come mainly from external factors such as increasing (international) competition. Therefore, it helps to understand how these companies face competition and which percentage of their sales is based on international sales. First, on average the companies sell 85.66 percent (n = 177, SD = 27.04) of their products to national consumers. Even 65.0 percent of the companies do not
Country Country education (%)
and year of measurement
Number of companies and percentages Albania 12.90 (2012) 11 (6.1%) Belarus 51.66 (2009) 9 (5.0%) Bulgaria 21.51 (2012) 7 (3.9%) Croatia 18.29 (2011) 23 (12.8%) Czech Republic 17.34 (2012) 7 (3.9%) Estonia 36.04 (2012) 13 (7.3%) Hungary 21.01 (2012) 12 (6.7%) Latvia 27.29 (2012) 13 (7.3%) Lithuania 29.33 (2012) 12 (6.7%) Moldova 18.86 (2012) 30 (16.8%) Poland 21.79 (2012) 12 (6.7%) Romania 13.25 (2012) 21 (11.7%) Slovak Republic 17.17 (2012) 9 (5.0%) Total 179 (100%)
sell products to foreign countries at all. Moreover, when asking the companies where the main market for its main product resides, 48.9 percent of the firms indicated that their main market is based in the same municipality where the firm is located. Of the remaining firms, 38.2 percent sells its main product to the national market, and only 12.9 percent sells its main product to the international market. Therefore, it can be concluded that for these companies, the international activities are not major yet.
Regarding competition, 105 companies indicated that they face on average competition of eleven competitors (n = 105, M = 11.72, SD = 16.40), however, the variation is major, as 52.4 percent of these 105 companies only faces competition of six companies or less. Moreover, next to these 105 companies, 42 companies indicated that the amount of competitors were too many to count.
With reference to the sales performance of the firms within this study it can be observed that over the last three years 28.5 percent of the firms were not able to increase their sales performance and got a decline in their sales. For 5.0 percent of the firms their sales remained the same, and the other 66.5 percent of the firms were able to have a sales increase. These increases in sales vary greatly between €690 and €24.1 million.
Building on those descriptive statistics, it is possible to proceed with the analysis and investigate whether hypothesis 1 and hypothesis 2 can be confirmed. In the next chapter the hypotheses will be tested and discussed to examine whether they are in line with the existing literature.
Main market for main product Percentage
Local 48.90%
National 38.20%
International 12.90%
5. RESULTS
This section elaborates on the main findings of this study and investigates whether the stated hypotheses should be accepted or rejected. By means of those investigations it will be possible to answer the main research question, which will accordingly be discussed in the conclusion.
Main results
Hypothesis 1
To test whether innovation activities generate higher sales when firms have a higher education level in their country, a moderator analysis is the appropriate method to use. The results of this moderator analysis are presented in table 8 on page 32. First of all, in model 1, it can be observed that the control variable, firm size, when investigated solely, has a significant impact on the results (B = 792.70, p = 0.011). When taking a look at the second model, it can be observed that the variables overall innovation activity (B = 613.77, p = 0.015) and education level (B = 752.13, p = 0.003) both positively and significantly influence the sales performance. Moreover, the control variable firm size has lost its significance. The last and third model includes the interaction effect of education level on the relation between innovation activities and sales performance. This model shows that there is a significant interaction effect of average education level on overall innovation activity and the firm its sale performance (B = 452.06, p = 0.028). Moreover, the adjusted R2 is highest in model 3 (0.117), which indicates that this model predicts the sales performance most accurately, compared to the other two models. Therefore, it can be concluded that hypothesis 1 can be accepted: innovation activities indeed generate significant higher sales when firms have a higher education level in their country.
Hypothesis 2
First, for testing the separate innovation activities and its impact on sales performance, four one-way ANOVAs are executed. These tests, which are presented in table 9 on page 32, result in three significant results: process innovation, organisational innovation and marketing innovation all three show positively significant outcomes. This means that there is a significant difference between the sales performance means of groups that do execute the specific innovative activities and sales performance means of groups that do not execute these separate innovative activities. As can be seen in table 9 on page 32, executing organisational innovation activities, leads to the highest average sales performance, namely a €2,083,270 increase. Solely focusing on product innovation does not lead to a significant increase in sales performance.
Next, the interaction effects with two innovation activities are analysed. Marketing and organisational innovation give the highest average increase in sales performance (M = 2544.46) and product and process innovation give the lowest (M = 1382.02). However, none of the combinations of two innovative activities showed up as significant. The tables with the means and p-values can be found in appendix C on page 47. Therefore, it can be concluded that a combination of two simultaneously executed innovation activities will not significantly contribute to a higher sales performance.
Discussion of findings
The findings of this study provide several insights that are discussed subsequently. First, based on the data and executed tests hypothesis 1 can be accepted. Education level indeed moderates the influence of innovative activities on sales performance. Since education level is an important determinant for human capital (Becker, 1962), this study extends studies that found positive relations between human capital within the firm, absorptive capacities and accordingly innovative activities (e.g. Alegre, Sengupta & Paiedra, 2013; Lund Vinding, 2006). However, these studies related to human capital specifically focused on knowledge within the firm, whereas this study focused on the average knowledge level of the country. Dakhli and De Clercq (2004) also took a country-based level for their measurement of human capital. They based human capital on three factors namely: educational attainment (average years of schooling and literacy rate), average income and life expectancy (Dakhli & De Clercq, 2004). The findings of their study imply that human capital positively influences the number of patents, expenditures for R&D and high-technology export; all based on a country-level measurement. Even though the outcome of that study is based on a different measurement, both studies amplify the importance of human capital for successful innovative activities.
However, not all studies related to human capital and innovation found positive results. This study did not distinguish between radical and incremental innovations. Nonetheless, several studies found contradicting results for human capital and its impact on radical innovations. Marvel and Lumpkin (2007) found that formal education is positively associated with innovation radicalness. Nevertheless, Subramaniam and Youndt (2005) found that human capital itself was negatively associated with the capability to innovate radically. Only when the individual knowledge gets shared with others, it will positively influence radical innovations (Subramaniam & Youndt, 2005). The effect of sharing knowledge with others is not explicitly studied in this research, however, as Subramaniam and Youndt (2005) show, social capital may be extremely critical to achieve the benefits of high human capital for successful innovative activities.
literature suggested that the four innovative activities are all increasingly important for firms within the food-manufacturing industry, there was no consensus on which combination of activities would contribute the most to increased sales performance. However, the outcome of this analysis can be reasoned logically. Product and marketing innovation both have a direct impact on the presentation of the (new) product, while process innovation may help to deliver this product more quickly and efficiently (Naidoo, 2010; Schilling, 2013). Thus, all three directly influence the sales process of products. Whereas, organisational innovation, will probably be the innovative activity that has the least direct impact on the sales of new or improved products. Moreover, organisational innovation was the activity that was executed least by companies in this sample. Only 38 companies within the sample set conducted organisational innovation. Therefore, their impact may be the lowest when interacting with the other innovative activities.
6. CONCLUSIONS
Based on all the provided literature and analyses, this section will provide a conclusion on the findings and main results. First a brief summary of the findings will be given. Then the theoretical implications will be presented followed by the managerial implications. Lastly, the limitations and avenues for further research will be discussed.
The goal of this study was to answer the research question: Do the innovation
activities of companies in the food manufacturing industry generate higher sales performance when influenced by a higher country education level?
Based on the analyses within this study it can be concluded that innovation activities of companies in the food-manufacturing industry indeed generate higher sales performance when influenced by a higher education level of the country. Previous studies already found results pointing in the same direction, however, with different measurement levels or different variables.
Moreover, as has been discussed in the literature, innovation activities for firms within the food-manufacturing industry should involve the four innovative activities as has been described in this study. However, this study finds that, for the food-manufacturing industry, focusing on the combination of four innovation types simultaneously does not lead to the highest significant increase in sales performance. Including product, process and marketing innovation simultaneously will significantly lead to the highest increase in sales performance for firms within the food-manufacturing industry. When firms do solely have budget to focus on one innovation activity, it will be beneficial to focus on either process, organisational or marketing innovation as these three all solely positively influence sales performance as well. Among them, organisational innovation gives the highest increase in sales performance, followed by marketing and process innovation.
Theoretical implications
Moreover, this study shows that it is beneficial to split innovation activities into more categories, since they impact sales performance in distinctive ways. These results may be of high interest for strategic management and innovation related literature.
Referring to the resource-based view it can indeed be argued that a sole focus on resources within the food-manufacturing industry may not be sufficient. This study amplifies the importance of dynamic capabilities by showing that countries that have higher average education levels have firms that are better able to turn innovative activities into higher sales performances. As identified by Cohen and Levinthal (1990), one of the most important dynamic capabilities for successful innovation are related to knowledge management and human capital of the firm its employees, and thus also education level. Knowledge management within companies has recently gained more attention in relation to innovation and firm performance (Alegra et al., 2013; Dalkir, 2013; López-Nicolás & Merono-Cerdá, 2011; Manlio Del Guidice et al., 2014). However, this study extends those fields of research by exclusively focusing on the human capital, and specifically education level, as an important part of knowledge management. Moreover, this study used human capital as a macro indicator, namely average tertiary education level per country. This means that this study took it from a broader perspective, as human capital specifically per firm was not taken into consideration. Thus the results of this study are relevant for strategic management and innovation literature, since it shows the importance of the overall knowledge and human capital of a country to simulate successful innovation activities. Hence, this study emphasizes that dynamic capabilities should not only be developed internally within firms, but can also be stimulated on a nation-wide level.
Managerial implications
Next, this study also shows firm managers how important it is to distinguish clearly among innovation activities and devote the right resources to stimulate the specific innovative activities. If firms do not have sufficient resources to exploit more than one innovative activity, it will be most beneficial for them to focus on process, organisational or marketing innovation. Nevertheless, when firms have sufficient resources to expand their innovative activities, it may be wise to devote these resources to product, process and marketing innovation. These three innovative activities will together lead to the highest significant increase in sales performance.
Limitations and avenues for further research
As yields for all studies, this study comes with several limitations and avenues for further research. These limitations and possibilities for further research will be briefly discussed below.
First, this study relies on a database gathered by the European Bank for Reconstruction and Development in partnership with the World Bank, which means that the data is not gathered for the sole purpose of this study. This makes it difficult to get an objective view of how the interviews precisely went and how accurate the measurements are. Nonetheless, the above stated institutions can be regarded as trustworthy institutions with the right intentions, which make it justifiable to use their dataset. Moreover, executing the survey solely for this research would have been impossible within the time and cost limits.
Secondly, Central and Eastern Europe has been chosen as a focus point of this study. However, even though these countries have a lot of similarities regarding for instance the entrance of the Single European Market and privatization, there are also still differences between these countries that may impact the results of this study. Therefore, a next study related to this topic may try to set up a survey and database that incorporates more (control) variables, that may explain the differences between the countries.
Thirdly, focusing solely on the food-manufacturing industry and on Central and Eastern Europe may decrease the generalizability of the study. For this study, these two factors: the food-manufacturing industry and Central and Eastern Europe were a well-chosen scope. Yet it may become interesting to increase the scope of this study to other industries and countries to see whether the outcomes are the same for other industries and regions.
sales outcomes of the company. Accordingly, Cordero (1990) argues that sales are not a real profit measure because it does not include resources used for the commercial units. Additionally, the sales indicator does not incorporate shocks in the economy and other socio-economic factors that may influence the (investment) decisions of the firm and thus the firm performance. However, for this research no right data concerning those factors was available. Moreover, as has been argued before, for innovation the most important outcome is successful commercialization, which is best measured by increased sales (Adams et al., 2006; Schilling, 2013). Nonetheless, it may be interesting to execute a similar study with different performance measure outcomes, such as return on investments.
