Master’s Thesis
MSc. in Business Studies – International Management
Knowledge-intensive manufacturing firms and the impact of
knowledge on their backsourcing decision
Yorick de Waardt - 10507000
University of Amsterdam - Amsterdam Business School First Supervisor: Lori Divito
Second Supervisor: Johan Linduque January 2014
Table of Contents
Introduction ... 4
Literature review ... 7
Backsourcing ... 9
Knowledge and Intellectual Capital ...13
Hypotheses ...20
Conceptual Framework ...23
Data and Methodology ... 23
Research design ...23
Data collection ...27
The VAIC model ...28
IC performance measurements ... 31 Dependent variables ... 32 Independent variables... 33 Control variables ... 33 Regression model ... 33 Results ... 34
Descriptive statistics and correlation analysis ...34
Regression analysis and the IC performance ...35
Discussion ... 38
Conclusion ... 43
References ... 46
Appendix I – List of Abbreviations ... 50
Appendix II – Final data set – SPSS input ... 51
Appendix III – Outputs from SPSS ... 54
Abstract
This study has investigated a possible relation between knowledge in manufacturing firms and the strategic decision to backsource or not their manufacturing activities back to the USA. For the hypotheses development, the Intellectual Capital, measured with the Value Added Intellectual Capital, and two individual elements, the Value Added Human Capital and Structural Value Added, have been used. Due to its simplicity and the reliable data input, the Value Added Intellectual Capital model introduced by Pulic (1998) has been employed to analyse the efficiency of knowledge within a firm. Contrary to the resource-based view that posits that intellectual capital and knowledge are competitive advantages and strategic assets, our study has generated mixed results. The regression analysis does show that the Value Added Human Capital generates “most” significance of all the variables in the Value Added Intellectual Capital model; instead, no significant relationship between the Value Added Human Capital and the backsourcing decision has been found. Neither is the Structural Value Added an element of any significance in the relationship between the Value Added Intellectual Capital and the backsourcing decision. Previous studies using the Value Added Intellectual Capital model have shown that the model is widely used in academia and as well as in practice. Nevertheless, Suggestions for further research include an improved analysis tool enriched with quantitative and qualitative elements is argued in the present paper to be able to bring new insights on whether or not more tactic or implicit knowledge within a firm is one of the drivers behind the backsourcing decision of manufacturing firms.
Introduction
The financial crisis started in late 2007, but one can already see the light at the end of the tunnel in America. The recession in Europe and Asia is stabilizing, and it is estimated that the dark clouds full of rain and thunder will disappear in the near future. Since the demand for goods in general, in combination with rising oil prices, has decreased, firms around the world have had a troubled couple of years, where cost pressures have been on the agenda of board members. The production of goods has become more expensive, even in the so-called “low-cost regions” like Asia or Eastern Europe. Rising labour costs, transportation costs, and quality control costs are just a few among other reasons why manufacturing firms are reallocating their facilities. Due to the decline in demand, boards had to decrease their manufacturing costs in order to rescue their firms from bankruptcy and to prepare the companies for better days, should these appear on the horizon. During the last five to six years it became clear that not all outsourced manufacturing facilities were a success. The combination of cost pressures from the firm boardrooms and rising costs on a local basis made some direct foreign investments not as successful as it had been hoped for. Even before the financial crisis firms had realised that not every outsourced activity had been necessarily successful. Rising costs for manufacturing firms are among main motives for backsourcing (Kinkel, 2012).
Van der Lee (2013) reports a reason why an outsourced activity could fail besides the rising costs, “the inability of firms to reduce Liability of Foreignness (LoF) in combination with inability to reduce costs are likely to choose a backshoring strategy” (Ibid., p. 2). LoF has been thoroughly investigated by numerous researchers
who describe it as “all additional costs a firm, operating in a market overseas, incurs that a local firm would not incur” (Zaheer, 1995, p. 343).
A deeper look into backsourcing might shed more light on whether a more knowledge-intensive or a relatively higher-level of intellectual capital in a firm has an impact on the strategic decision whether to backsource or not. Intellectual capital is commonly defined as the sum of all knowledge resources a firm can use in order to generate a strategic competitive advantage (Nahapiet & Ghoshal, 1998; Young, Su, Fang, & Fang, 2009). A higher-level of intellectual capital or knowledge level within a firm is the sum of all kinds of knowledge compared to other backsourcing and non-backsoucing manufacturing firms, this includes tactic or implicit knowledge and explicit knowledge. Tactic knowledge is not easily transferable and can only be learned through intensive interaction with other persons, where explicit knowledge can be obtained through books and is based on facts (Polanyi, 1966).
By looking at how knowledge-intensive a firm is in the year prior to its backsourcing decision, we might be able to see if a higher level of intellectual capital is the reason for a firm to backsource. A knowledge-intensive firm is an organisation where the sum of all knowledge resources a firm can use to generate a competitive advantage is relatively high (Starbuck, 1992; Young et al., 2009). While ‘knowledge-intensive’ can be defined in numerous ways (Rylander & Peppard, 2009), in this study, the following definition will be used: “knowledge-intensive firms refer to those firms that provide intangible solutions to customer problems by using mainly the knowledge of their individuals” (Ditillo, 2004, p. 401).
In the future firms that consider backsourcing could look at the knowledge intensity or intellectual capital as used in this study and see whether these match their own firm characteristics.
The main purpose of this study is to get an insight on whether or not knowledge, or intellectual capital, is one of the drivers behind the decision of a firm to backsource. The method section is based on the previous research on intellectual capital by Chen, Cheng, & Hwang (2005) and Firer & Williams (2003). A total of 119 manufacturing firms from the Standard & Poor 500 (S&P 500) were analysed with the Value Added Intellectual Coefficient model from Pulic (1998). Because the outsourcing and backsourcing trends originate from the USA, this study limits itself to the investigation of US firms only. Data on backsourcing and non-backsourcing manufacturing firms are collected through the S&P 500 database. These data are reliable and standardised through the single accounting standard used in the USA, the General Accepted Accounting Standards. The Value Added Intellectual Coefficient is the sum of three individual elements of the model: human capital efficiency, structural capital efficiency, and relational capital efficiency. The human capital efficiency and the structural capital efficiency are the dimensions which are individually analysed in their relation to the backsourcing decision.
The structure of this paper is as follows. First, a literature review providing a detailed background for this empirical study is presented and followed by the
conceptual framework and the methodology sections. Furthermore, the findings of this study are presented in the results, discussion, and conclusion sections. A critical view on this research will be presented in the discussions part, together with the recommendations for further research.
Literature review
Outsourcing
Outsourcing started in the 1960s in the USA when a transistor manufacturing company first outsourced its activities to Mexico. Outsourcing was not formally identified as a business strategy until 1989 (Mullin, 1996). Eastman Kodak’s decision to outsource their information technology systems was quite revolutionary in 1989, and it was the real start of the outsourcing activities. In 1970, there were already firms outsourcing their activities, but those activities were more concerned with external acquisitions rather than with outsourcing proper. As outsourcing was becoming known and grew in popularity, the question gradually changed from whether or not to outsource to how much to outsource (Lee, Huynh, Kwok, & Pi, 2003).
Among the previously reported reasons for and advantages of outsourcing is that when organisations outsource parts of their in-house operations, significant savings on operational and capital costs could be gained (Hendry, 1995; Rimmer, 1991; Uttley, 1993). This can allow firms to focus on their core activities (Arnold, 2000; Hendry, 1995; Prahalad & Hamel, 1990). Other studies observe that, as suppliers may be significantly more advanced in their technology, outsourcing allows organisations to exploit their more advanced knowledge (Greaver, 1999). However, these findings pertain to the previous century; nowadays, the views on outsourcing have changed significantly, and, due to the nature of outsourcing activities, not all of them can be directly applicable anymore.
There are several frameworks that thoroughly explain the outsourcing decision and show its advantages that could be generated in the different markets (Peng, 2012). The internationalisation theory (Buckley & Casson, 1976; Rugman & Hodgetts, 1995)
and the paradigm of ownership, location, and internationalisation (OLI) advantages (Dunning, 1980, 1988) are the commonly used theories. The internationalisation theory anticipates that the production in the home country will specialise in the production elements where skills are of importance together with capital intensive elements of the firms. The activities at the foreign location of the firms will be able to take advantage of the low costs at the host country locations where more labour intensive elements of the production process is located (Dunning, 1980, 1988). Therefore, the main reason behind the outsourcing decision is that a firm can expand or transfer its operations by going abroad and try to generate more efficient and effective ways of doing business. The low labour costs are one of the most used arguments for outsourcing standardised manufacturing operations. Standardised manufacturing operations are the production processes of physical goods through the use of labour and machinery with the purpose of usage or sales.
Apart from the internationalisation theory where location is one of the main drivers of the theory, there are other models which explain how firms can generate more efficiency by looking at the tangible and, more importantly, the intangible resources of a firm. The resource-based view (RBV) is a framework that can explain the strategic outsourcing or backsourcing decision, as it can explain why firms switch the location of their manufacturing operations. A firm might have different motives to reallocate its production. Resource seeking, market seeking, efficiency seeking, and strategic asset seeking are used in the RBV model as the four main arguments for the reallocation of activities (Peng, 2012). The RBV model is linked to the ability of a firm to generate a competitive advantage though the tangible and the intangible capabilities and assets. In order to create a competitive advantage through a sustained
competitive advantage, it needs to be heterogeneous and not easily imitable. The capabilities and assets could then turn into valuable resources that are neither perfectly imitable nor substitutable without great effort. Together with these dimensions and the fact that the resources give the firm an above-average return, it can be described as a sustained competitive advantage. According to the RBV, intangible resource is the most valuable resource a firm can possess, as it can give the firm an above-average return and, therefore, a sustained competitive advantage (Peng, 2012). The specific knowledge held within a firm is as such an intangible resource. This specific knowledge can generate a sustained competitive advantage and an above- average return.
Backsourcing
There are numerous definitions of backsourcing in the literature. Although it has variously been termed insourcing, inshoring, and reshoring, all the definitions have one common description — the activity of bringing back production to the home location that has been outsourced in the past. In this study, the term backsourcing will be used to refer to the activities of bringing the production back to the home location, i.e. the location where the production took place before it was outsourced.
The motives for the backsourcing of the outsourced activities are quality problems, loss of flexibility, rising labour costs, and high coordination and control costs (Kinkel, 2012). A study by Kinkel and Maloca (2009) concludes that the outsourcing trend has lost momentum. By studying 1663 German manufacturing firms on the offshoring and the backsourcing activities, the authors show that every fourth to sixth offshoring activity is followed by a backshoring activity in the next
four years, mainly due to the lack of flexibility and quality problems at the foreign location.
The study by Kinkel & Maloca (2009) is focused on German firms in the industrial sector, and, although the industrial sector in Germany is quite extensive (IMF, 2012), a limitation of this study is its focus on one country only, which raises doubts whether these findings can be used at a broader international level. As one of the authors himself admits, “Thus, backshoring of manufacturing capacities might be a quantifiable phenomenon, but reliable data are not yet diffused in academic discussion.” (Kinkel, 2012, p. 697). However, a study by Kinkel (2012) mentions that backsourcing as an initiative to bring back the production facilities to the home country is growing not only in Germany, but also in the USA where the backsourcing trend initially started.
Kinkel (2012) looks at the manufacturing industry in Germany in two studies, one related to manufacturing firms, and the other focused on the automobile industry. In the literature review on the trends in reallocation and backsourcing activities, the author concludes that the only available European dataset on reallocation and backsourcing activities was used in both cases, “The written survey set has been carried out by the Fraunhofer Institute for Systems and Innovation Research (ISI) every two years since 1995. It is the only dataset in all of Europe which regularly enquires about the trends towards relocation as well as backsourcing of production and R&D activities in manufacturing industry” (Ibid., p. 701). As mentioned above, in his study on the German automobile industry he refers to the same data set, but uses a different time-frame.
In line with Kinkel’s (2012) study and other research on the backsourcing phenomenon since 2010 (e.g., Mcmeekin & Mcmackin, 2012; Mondal, 2011), it can be argued that backsourcing has become a trend. Reputed firms in the consultancy branch, such as the Boston Consulting Group or Accenture, have invested time and effort into the analysis of the backsourcing trend. “Within the next five years, the United States is expected to experience a manufacturing renaissance as the wage gap with China shrinks and certain U.S. states become some of the cheapest locations for manufacturing in the developed world” (Fondiler, 2011). Most of the backsourcing activities are realized in USA because the States have the highest percentage of outsourced activities to China (Harrington, 2011).
According to the study by Accenture, there are certain elements that are most relevant for the backsourcing decision. 61% of respondents are currently considering shifting their manufacturing operations closer to the customers to provide better service and to accelerate their company’s growth. Other reasons for pulling the activities back include cycle/delivery time (49%) and product quality (49%). The control costs of the manufacturing firms are getting higher so that it becomes more interesting to produce in-house again. Furthermore, for Asian manufacturers during 2007-2010, logistics (57%) and supplier/component prices (73%) were most frequently mentioned backsourcing reasons. Notwithstanding, the most important factor in rethinking the outsource strategy is the rise in labour costs (74%) (Ferreira & Meilala, 2011). All of the above are important reasons to re-consider if the outsource activity is still in the best interest of the firm. If the backsourcing decision is really a significant improvement, why do not all the manufacturing firms backsource their outsourced activities?
This could be explained by the fact that the backsourcing strategy requires the recruitment of new employees, which is complicated, costly, and generates no guarantied satisfaction (Chapman, Andrade, & Anadle, 1997). Whitten and Leidner (2006) suggested that, “the results indicate that product quality, service quality, relationship quality, and switching costs are related to the decision to backsource application outsourcing, with product quality and service quality being the differentiators in deciding whether a client will backsource or switch.” (Whitten & Leidner, 2006, p. 617). In the end, only five per cent of companies achieved significant benefits from outsourcing (Lonsdale, 1999). There are several other empirical studies on the backsourcing decision, but most of them are related to IT and information systems, rather than production or manufacturing operations (Overby, 2005; Whitten & Leidner, 2006).
Two case studies carried out at the firms from the United Kingdom and USA provide a detailed analysis of the strategies to follow in case a firm is considering backsourcing and wants to ensure an effective knowledge re-integration of the backsourced activities. The study by Bhagwatwar, Hackney, and Desouza (2011) describes several steps that should be taken into consideration in order to ensure that the knowledge gained during the outsourcing period is re-integrated when the operations are backsourced. The study reports on the experience of two firms that have learned from their outsourcing decisions and have gained new knowledge about the reasons why the outsourcing decision was not the right strategic direction after all. This study provides an important example of the gained knowledge and discusses the reasons why this gained knowledge is pivotal in the decision for or against backsourcing. If two competitors have outsourced their activities to the same low-cost
production region, specific knowledge on, for example, costs, cost reduction opportunities, or market demand shifts is a valuable asset. “Firms compete on the basis of the superiority of their information and know-how, and their abilities to develop new knowledge by experiential learning.” (Kogut & Zander, 1993, p. 640). Armed with superior knowledge, one can make a certain strategic decision, be it an outsourcing or a backsourcing decision, before their competitors can.
Knowledge has an intangible value and, according to the RBV (see above), it is the most valuable asset a firm can have (Ahmad & Mushraf, 2011). Similarly, it has also been claimed that knowledge is our most powerful engine of production (Nahapiet & Ghoshal, 1998).
Knowledge and Intellectual Capital
In this study, knowledge and knowledge generation within a firm, as well as the reasons why it creates a competitive advantage are considered relevant because knowledge does not appear out of nothing. The process of creating knowledge within a firm has been studied first by Polanyi (1966) where the epistemological dimension of knowledge has been introduced. The epistemological dimension of knowledge identifies two different elements: tacit knowledge and explicit knowledge. The work by Nonaka (1995) reintroduced the knowledge literature by designing a tool to represent the transfer of knowledge within a firm, and, ever since then, this tool has been used extensively by researchers and practitioners over the years.
Tacit knowledge is more related to the person and is therefore more subjective, it is the so-called “know-how”. This type of knowledge is generated through experiences and cultural behaviours, which makes it not easily transferable.
Learning experience and thus knowledge generation are solely possible through interactions (Nonaka, 1995). Explicit knowledge is more objective and is based on facts; therefore, it can be acquired through books or other learning assistance and is thus easy transferable (Nonaka, 1995).
The functions of knowledge within firms has been studied extensively by Kogut & Zander (1992, 1993, 1995). The study from 1992 demonstrates a static dimension associated with knowledge or intellectual capital, this dimension will lead to the creation of new market opportunities. Another element this research elaborates on is the dynamic perspective of knowledge where the resources and capabilities of a firm come together in order to establish knowledge creation and an organizational learning experiences (Kogut & Zander, 1992). In this study, the static dimension is the main focus.
The concept of ‘intellectual capital’ (IC) can broadly be defined as the knowledge within a firm. Some practitioners argue that IC is nothing more than a term that economists and accountants use to refer to knowledge within a firm (Denning, 2009). The term IC allows economists and accountants to differentiate between intellectual capital, equivalent to knowledge, and information. Zander & Kogut (1995) differentiate between ‘know-how’ and information, where ‘know-how’ is the “knowledge held by individuals, […] expressed in regularities by which members cooperate in a social community” (Marr, Schiuma, & Neely, 2004, p. 774), and information is data formatted in such a way that they represent a meaningful pattern. In practice, there is no a clear distinction between knowledge and the IC, as both can be considered to be two sides of the same coin.
Denning (2002, p. 33) concludes with a clear example between information and IC, or knowledge:
“A notorious example of the confusion is the opening statement of the World Development Report on Knowledge for Development (1998-1999), a document written principally by economists, which begins with the extraordinary and false assertion that knowledge travels at the speed of light. In fact it can be extremely easy and quick to transfer information from one place to another, intellectual capital (knowledge) is often very difficult and slow to transfer knowledge from one person to another.”
IC is defined by Stewart (1997) as an “aggregation of all kinds of knowledge and competences of employees, which could bring about competitive advantages for firms” (Sharafi, Mohammadi, & Sharafi, 2012, p. 148). Other established researchers in the field define IC as “the possession of knowledge, applied experience, organizational technology, customer relationships and professional skills that provide a competitive edge in the market” (Bontis, 1999; Stewart, 1997). Considering that there are multiple similar definitions, in this study IC will be assumed to refer “the knowledge and knowing capability of a social collectivity, such as an organization, intellectual community, or professional practice” (Nahapiet & Ghoshal, 1998, p. 245). According to Bontis (1998) and other researchers in the IC area, IC consists of three interrelated elements, namely: human capital, structural capital, and relational capital. Human capital, as shown in Table 1, comprises human resource elements, such as attitudes, employee competencies, experience and skills, tacit knowledge, and the innovativeness and talents of employees (Chen et al., 2004; Choo & Bontis, 2002;
Guerrero, 2003; Roos & Jacobsen, 1999). Human capital (HC) is a crucial element in a firm for innovative and strategic renewability (Wilson & Larson, 2002). At the point where new employees with certain knowledge, which is not yet owned by the firm, get recruited by the firm, it will increase that firm’s human capital (Grasenick & Low, 2004). Conversely, when employees leave the firm, they withdraw their gained competencies, tactical and strategic knowledge elsewhere, so that the firm loses its human capital (Bontis, 2001; Edvinsson, Roos, Roos, & Dragonetti, 1997; Grasenick & Low, 2004).
Table 1 – The indices of human capital
Employees competence - Strategic leadership of the management - Qualities of the employees
- Learning ability of the employees - Efficiency of employee training
- The employees’ ability to participate in policy making and management
- Training of key technical and managerial employees Employees attitude - Identification with corporate values
- Satisfaction degree - Employees’ turnover rate
- Employees’ average serviceable life Employees creativity - Employee’s creative ability
- Income on employees’ original ideas
(Chen et al., 2004)
Structural capital (SC) (see Table 2) is the second element in IC, and consists of the supportive infrastructure and knowledge within a firm such as processes, databases, routines, strategies of the firm, and the specific culture within the firm in order to let the human capital side of the firm work properly (Bontis, 2001; Chen et al., 2004).
Table 2 – The indices of structural capital Corporate culture - Construction of a company’s culture
- Employees identification with a company’s perspective Organizational structure - Clarification of relationship among authority,
reponsibility, and benefit
- Validity of enterprise controlling system
Organizatinal learning - Construction and utilization of inner information net - Construction and utilization of company repository Operation process - Business process period
- Product quality level
- Corporate operating efficiency
Information system - Mutual support and cooperation between employee - Availability of enterprise informatioin
- Knowledge sharing (Chen et al., 2004)
The third and last component of IC is relational capital (RC), which relates more to the external relationships, formal and informal, that a firm can have. It is the opinion which external stakeholders have of the firm, together with the interchange of certain knowledge between the firm and the external stakeholders (Bontis, 1998, 2001; Chen et al., 2004). Examples include the loyalty and the preferences of the stakeholders, the reputation of a firm built through the years, the interaction with the local government, the established network, and the information gained about competitors. Table 3 represents only the consumer-related elements of relational capital (after Chen et al., 2004). The terms ‘consumer capital’ and ‘relational capital’ are sometimes used interchangeably in the IC-related papers, which can be confusing (Bontis, 1998; Maditinos et al. , 2011). Chen and co-authors (2004) include only the elements related to consumers, which is a limitation in their research. In the present study, the term ‘relational capital’ will be used instead.
Table 3 – The indices of customer capital
Basic marketing capability - Construction and utilizationof the customer database - Customer service capability
- Identifying ability of customer’s needs
Market intensity - Market share
- Market potential - Unit sales to customer
- Brand and trademark reputation - Construction of sales channel Customer loyalty indices - Customer satisfactioin
- Customer outflow
- Investment on customer relationship (Chen et al., 2004)
Looking at IC in combination with the RBV we can conclude that IC is a valuable asset to any firm considering that IC is knowledge and it has intangible nature. The RBV states that, in order to create a competitive advantage through a sustained competitive advantage, it needs to be heterogeneous and not easily imitable. The capabilities and assets could then turn into valuable resources that are neither perfectly imitable nor substitutable without great effort. According to Peppard & Rylander (2001), the IC literature and the RBV are almost inseparable. It must be mentioned however that IC is highly contextual and not all investments in IC would have the same outcome for different firms (Bontis, Dragonetti, Jacobsen, & Roos, 1999; Roos & Jacobsen, 1999; Snyder & Pierce, 2002).
Over the years, IC and its components have been extensively analysed by researchers in different industries. Most of this research has been focused on whether IC has a positive or negative influence on a firm performance, and which of these individual components have more impact. A study on 96 Greek companies shows that there is no significant support for the conclusion that IC impacts their financial
performance. At the same time, this study has not found a positive relationship between HC and the financial performance of the firms (Maditinos et al., 2011). This result coheres with the previous research by Young, Su, Fang, & Fang (2009) which focused on eight Asian economies. After controlling of the influence of the financial crisis, the results show that HC is the major element generating value and higher performance for banks.
Other results on IC and its individual elements have been reported in the research on IC as concerns its influence on the stock market performance of a firm. In the high-tech industry, a positive relation has been established to exist between IC, financial performance, and the stock market performance. Further analysis shows that physical and financial assets within a firm are important influential factors in these financial and stock market performance measurements (Zeghal & Maaloul, 2010).
A different angle has been taken the study by Musibah, Sulaiman, Wan, & Alfattani (2013) which considers the relationship between IC and Corporate Social Responsibility (CSR) in the Islamic banking sector. The IC is shown to have a strong negative impact on CSR; however, the size of a firm might make it more difficult for firms to control their CSR.
In South Africa, available research does not generate evidence to support any relation between IC or its elements, on the one hand, and corporate performance, on the other hand. One of the relevant studies analyses 75 listed firms and reports a mixed outcome in the overall generated results; only some of the results substantiate a moderate positive relation between the physical capital within a firm and corporate performance (Firer & Williams, 2003).
The manufacturing sector of Taiwan has been investigated for a relationship between IC, the firm market value, and financial performance, and a positive relation between these variables was established. Together with the physical capital of manufacturing firms, HC, as one of the individual elements of IC, has been shown to be of a great importance in this positive relation (Chen et al., 2005).
The main findings from the studies overviewed above are that there is a positive relation between Value Added Human Capital (VAHC) and firm performance in Greece, all other elements of the Value Added Intellectual Capital (VAIC) model where inconclusive. In eight Asian economies, a relation has been found between VAHC and the financial and stock performance of banks. In South Africa, no relation has been reported between IC, or its individual elements, and the performance of a firm. Only in Taiwan a positive relation between IC and financial and stock performance has been found, and VAHC has been shown to be a major element contributing to this relation. VAHC is one of the elements of the VAIC model that has been shown to contribute most to a tentative relation between IC and a performance measurement.
Hypotheses
Consistently with the previous research overviewed in the preceding section, it can be expected that IC within a firm can be positively related to a firms’ financial and stock market performance (de Waardt, 2012). IC is closely related to the RBV of a firm, and the RBV states that intangible resources are the most valuable resources a firm can possess, in that they can give the firm an above-average revenue and, therefore, a
such an intangible resource. Considering the backsourcing trend and the fact that a backsourcing firm experiences a positive increase in the performance, the following hypotheses can be formulated:
H1a: Firms with a greater Intellectual Capital are more likely to backsource.
Previous studies also showed mixed results on the relationship between the individual elements of IC and the performance of the firms. Thus, it is important to see whether one individual element, or a combination of elements, can explain a possible relation between IC and the backsourcing decision. Human Capital is seen as a crucial element in a firm for innovative and strategic renewability (Wilson & Larson, 2002). At the point where new employees with the knowledge that is not yet owned by the firm get recruited by the firm, the firm will increase its human capital (Grasenick & Low, 2004). Conversely, when employees leave the firm, they take their gained competencies and knowledge elsewhere (Bontis, 2001; Edvinsson et al., 1997; Grasenick & Low, 2004). Firms with greater human capital will have more knowledge at their disposel, this gives the firm a better understanding of the market condisitions and a competitive advantage over their competitors (Peng, 2012). The backsourcing trend has shown that the decision to backsource manufacturing activities is a positive development for firm that outsourced their activities to low-cost regions (de Waardt, 2012). Therefore, the followig hypothesis can be formulated:
Structural capital is the second element in IC; it consists of the supportive infrastructure and knowledge within a firm including processes, databases, routines, and strategies of the firm, as well as the specific culture within the firm allowing the human capital aspect of the firm work properly (Bontis, 2001; Chen et al., 2004). The human capital, or employees within a firm, could simply not work without this supportive infrastructure. Greater structural capital will increase the support to the employees, if the employees can function better the human capital within the firm will increase. This leads to a higher level of knowledge and thus a better competitive advantage. Therefore, the following hypothesis can be formulated:
H3a: Firms with a greater Structual Capital are more likely to backsource.
In the present study, both backsourcing and non-backsourcing firms are used to analyse the relationship. In addition, taking into account the Heckmann correction for a selection bias, the following secondary predictions can be formulated (Heckman, 1979):
H1b: Firms with lesser Intelletual Capital are more likely not to backsource. H2b: Firms with a lesser Human Capital are more likely not to backsource. H3b: Firms with a lesser Structual Capital are more likely not to backsource.
Although IC consist of three interrelated elements, the impact of relational capital on the backsourcing decision is not taken into consideration, because it is an element which has its value outside of the backsourcing and non-backsourcing firms.
Conceptual Framework
Data and Methodology
Research design
There are multiple methods to measure IC in a firm based on monetary and non-monetary valuations (also called dollar valuation and indicator-based measurement tool). The developed non-monetary valuations are the Balanced Scorecard by Norton & Kaplan (1992), Skandia Navigator outlined in Edvinsson & Malone (1997), and Roos et al.'s (1997) IC-index. The models based on the monetary valuation methods include the Calculated intangible value method (Tseng & James Goo, 2005), the Economic Value Added tool (Bontis, 1999, 2001; Stewart, 1997), and the Value Added Intellectual Coefficient (VAIC). The VAIC is a measuring tool whereby the total efficiency of human capital, structural capital, and capital efficiency is measured.
When a firm has a high VAIC, it is argued that the firm has “good management utilization of the potential value creation from physical and intellectual capital” (Tseng & James Goo, 2005, p. 192) This tool focuses on the outcomes of a project, and it can be used within multiple levels, countries, and businesses (Pulic, 2000).
In this study, the choice has been made to use the VAIC model to determine the IC value of a firm. Multiple researchers have used this tool and summarized the advantages of this model. There have been numerous studies that investigated the relationship between IC and firm performance. By looking at financial statements and public records, these studies were able to generate a quantitative database. This line of data gathering made it possible to use the VAIC method and analyse the creation of the IC efficiency. In what follows, we will briefly discuss previous studies on IC with the VAIC.
Chen, Cheng, & Hwang (2005) investigate the relationship between value creation efficiency, the market value, and firm performance in Taiwanese public firms. This study reports a positive relationship between IC as measured by the VAIC, on the one hand, and the market value and performance of a firm, one the other hand. A study by Zéghal & Maaloul (2010) on 300 UK firms has found supporting evidence for a positive impact of IC on economic and financial performance using the VAIC, and concluded that the VAIC is an important tool for many decision-makers. Tan, Plowman, & Hancock (2007) also investigate the link between IC and the performance of firms on the Singapore Exchange with the VAIC and find that IC and future firm performance are positively related. Firer & Williams (2003) studies 75 public firms in South Africa. Their results indicate that the physical resources in a firm are positively related to the corporate performance.
In Russia, Molodchik & Bykova (2011) have applied the VAIC in the study of 350 Russian firms and have found a positive relation between IC of the firms and their future competitiveness.
Corporate social responsibility (CSR) and the influences of IC within a firm are examined by Musibah, Sulaiman, Wan, & Alfattani (2013). The study uses the VAIC in gathering insight on CSR in Islamic banks of the years 2007–2011 and reports significant relations on multiple elements of the VAIC model.
In the hotel industry of Australia, a study has been conducted with the VAIC and the findings show that the VAIC model is a proper tool to analyse the efficient use of intellectual capital within a firm (Laing, Dunn, & Hughes-Lucas, 2010). In an analysis of eight Asian economies, another study shows that the physical capital and human capital are the elements generating most of the value for financial institutions. This study used the VAIC model to analyse firms from 1996 to 2001 (Young, Su, Fang, & Fang, 2009).
The reason why all the studies overviewed above use the VAIC is unanimous. The model created by Pulic has the advantage of using publicly available quantitative data (Pulic, 2000). By gathering the data from financial statements, one can generate the ratios needed. Other models use questionnaires and are limited to unique data fitting only a few, or only a single firm (Edvinsson et al., 1997). Other models have limitations because the data they use are not, or have not been recorded by the firms and are only retrievable through questionnaires. As a result, other IC measurement methods are not consistent, and the validity is threatened, which makes comparative studies impossible (Firer & Williams, 2003).
While the VAIC model also has its limitations, it is, notwithstanding, a simple, reliable, and comparable method (Maditinos et al., 2011). Andriesson (2004) concludes that making use of the VAIC as a measurement tool is legitimate due to the availability of public records. These financial statements are also audited, so the input for the VAIC model is standardized and verifiable (Firer & Williams, 2003).
A critical review on the VAIC model is based on the calculations of randomly selected firms from the Financial Times Stock Exchange 250 (FTSE 250), which is an index including one hundred and one largest firms on the London Stock Exchange. This paper examines the validity of the VAIC model and concludes that there is only one positive correlation between a component of the VAIC model and performance, measured by market to book value ratio, of the firms (Ståhle et al., 2011). As no correlation is found among the elements of the VAIC model, the authors fail to offer a satisfactory explanation to their finding, as they find the elements of the model to be not clear and not transparent enough. Two main arguments are stated in the conclusion of the study at stake: first, there is “a confusion of capitalized and cash flow entities in the calculations of structural capital” (Ståhle et al., 2011, p. 547) and, second, there is a “misuse of intellectual capital concepts” (Ibid.). The authors compare their study with the studies overviewed above (Lev, 1996; Pulic, 1998; Stewart, 1997), and also mention the studies with weak correlations between IC, measured with the VAIC model, and firm performance. Numerous studies that have used the VAIC in the past show that the model has its limitations, but these limitations do not outweigh the pro-VAIC model arguments. Previous studies show the strengths and the validity of the VAIC model as the measurement tool for IC.
Therefore, the VAIC model will be used in this study to examine IC in manufacturing firms in the S&P 500.
At this point it is important to mention that the VAIC model has a great explainatory power for the intellectual capabilities and the knowlegde level of a firm; hence, researchers argee that if different values are placed on the three elements of the VAIC model, the VAIC model will have a better explanatory power (Chen et al., 2005). If the model would solely use the sum of the three elements, the model would not be as informative.
Data collection
The data collected for this study consists of the records from the publicly-traded manufacturing firms in the Standard & Poor 500 (S&P 500) over the period of eight years, from 2006 until 2013. The S&P 500 is a stock market index that consists of 500 largest firms listed on the New York Stock Exchange (NYSE) or the National Association of Securities Dealers Automated Quotations (NASDAQ). Through Datastream, a financial database with all the available data from the publicly traded firms, the 2006-2013 fiscal year annual reports were collected. The collected data are limited due to the nature of this study and the fact that a homogeneous dataset is used (Firer & Williams, 2003). All manufacturing firms in the S&P 500 are gathered for a more comprehensive study of all firms who have or have not backsourced their production. Manufacturing physical products is the feature that has been used as the first criterion for selecting the firms. The other criterion has been that the manufacturing process or facility should have been able to be outsourced to another location abroad. Meeting these two criteria has been accomplished via considering the
company descriptions in the S&P 500 database. If the firm stated that it produced physical products in its own facilities, and that the production process was not bounded by the location, the firm has included in the non-backsourcing manufacturing list. As mentioned before, the firms that do not produce physical products, or the firms with the production facilities that cannot be outsourced, are not included in this study. The latter group of the firms that have not been included into our dataset can be illustrated by an oil production firm that produces oil in a certain region. Outsourcing the production of this firm to another location is not possible, as the production relies on the oil resources found at that specific location which are not available elsewhere. The included firms are all situated in the United States of America as their production facility, and have been, or are listed in the S&P 500. A number of backsourcing firms have been taken from our previous study on backsourcing and its relation to the stock price performance (de Waardt, 2012). Backsourcing firms were found via the google search engine and through various secondary data sources, such as newspaper articles or press releases (see Appendix III for further detail). After excluding firms with incomplete data, 108 non-backsourcing firms and 11 backsourcing firms remained in the dataset (see Tables 4 and 5 in section “Descriptive statistics and correlation analysis”).
The VAIC model
This research makes use of the VAIC model (Chen et al., 2005; Firer & Williams, 2003; Pulic, 2000; Young et al., 2009) to measure the knowledge within manufacturing firms. The VAIC model uses balance sheets and profit & loss accounts in order to calculate the needed ratios. When a firm has a high VAIC, it is argued that
physical an intellectual capital” (Tseng & Goo, 2005, p. 192). This tool focuses on the outcomes of a project, “its value added (VA) represents the value it produces through its resources/capitals” (Young et al., 2009, p. 1568). This value is measured by the output (OUT) of a firm and the input (IN) it uses, as shown in (1).
VA = OUT – IN (1)
The earnings of a firm are represented by the output. The inputs are all the expenditures a firm has apart from the employee costs. According to Pulic (1998), the employee expenditures cannot be seen as costs, as human capital is crucial in the process for the firm to create a long-term value later on. Employee expenditures should be seen as an investment, which can be withdrawn from the total expenditures. According to Chen et al. (2005), Firer & Williams (2003) , and Pulic (2000), the next step in calculating the VAIC is to calculate the capital employed (CE), the human capital (HC), and the structural capital (SC). The calculations for these elements are shown in (2)-(5).
CE = physical capital + financial assets (2) or
CE = total assets – intangible assets (3) HC = total expenditure on employees (4) SC = VA – HC (5)
The calculation for the total expenditure on employees is consistent with Chen et al. (2005) and Riahi-Belkaoui (2003). This variable can be calculated using the formula in (6):
W = S – B – DP – I – DD – T – R (6)
where W is the total expenditures on employees or wages; S is the total sales revenue;
B is the costs of the sold goods; DP is the depreciation; I is the paid interests; DD are
the yearly dividends; T are the taxes paid; and R is the difference between that year’s and the previous year’s retained earnings.
The RBV of a firm explains that the resources a firm has are the major contributors to its levels of competitiveness and performance, where the intangible assets are more valuable than the tangible assets. CE represents the tangible assets in the VAIC model, and HC is a proxy for the intangible assets (Chen et al., 2005; Riahi-Belkaoui, 2003).
In the next step, the VAIC itself can be calculated from the computation of its three elements: the value added of the capital employed (VACE), the value added efficiency of human capital (VAHC), and, finally, the value added of the structural capital (STVA) (Pulic, 1998; Young et al., 2009). The formulas are provided in (7)-(9).
VACE = VA ÷ CE (7) VAHC = VA ÷ HU (8) STVA = SC ÷ VA (9)
The VAIC is then the sum of these three elements (Pulic, 1998; Young et al., 2009, see (10)).
VAIC = VACE + VAHC + STVA (10)
IC performance measurements
In order to see if the backsourcing firms have a higher knowledge level, measured through the VAIC and its individual elements VACE, VAHC, and STVA, the first and the second year that preceded the backsourcing announcement within a firm is indicated by “– 1Y” and “-2Y”.
A higher-level of intellectual capital or knowledge level within a firm is the sum of all kinds of knowledge compared to other backsourcing and non-backsoucing manufacturing firms, this includes tactic or implicit knowledge and explicit knowledge. Tactic knowledge is not easily transferable and can only learned through intensive interaction with other persons, where explicit knowledge can be obtained through books and is based on facts (Polanyi, 1962).
The “0” thus is the year when the firm made its backsourcing decision public. For each year when the backsourcing firms made their backsourcing announcements (within the timeframe of 2009—2012), a randomly selected group of non-backsourcing firms has been extracted from the non-non-backsourcing firms group (N=108) in order to compare the backsourcing announcement in the year “0” with the year preceding the announcement year (i.e. “-1Y”). Each of the non-backsourcing groups contains one-fourth of the total number of firms in the non-backsourcing category. This randomly selected group has only been used for a specific year starting from 2009. If a non-backsourcing firm is used in the 2009 group, it is not be used in the group for 2010, 2011, and 2012. This has been done to ensure that a bias will not
occur through the use of non-backsourcing firms in multiple years. In sum, four groups have been generated (for the years 2009 – 2012). Merging these four groups has amounted to creation of a single group ordered not by years, but by the announcement year (“0”) and the two years preceding the announcement year (“-1Y” and “-2Y”). This new group has to be created for two reasons. First, not all of the backsourcing announcements were made in one year. Second, the non-backsourcing group is divided into four groups to ensure that in the merged group, which integrates the backsourcing and the non-backsourcing firms, the firms are only used once. If the non-backsourcing firms were used more than once in the merged dataset, it would contain a bias, because the dataset would then consist of more than 118 firms. If a firm were used twice, it would influence, negatively or positively, the outcome of the analysis. This has also to do with dependent variable and the logit regression analysis, explained more thoroughly below.
In what follows, we will present the results on IC or knowledge performance, to see if, in the year preceding the year of the backsourcing announcement, the backsoucring firms outperformed the non-backsourcing firms. These results will be followed by the results yielded by the regression analysis. A logit regression analysis will be used to see if the independent variables influence the dependent variables.
Dependent variables
In this study, a firm’s decision is categorized into two options: “to backsource” and “not to backsource”. These two outcomes are the dependent variables. The outcome whether or not to backsource is influenced by specific elements, the VAIC and its components, within the firm.
Independent variables
In this study, the VAIC and its three elements - VACE, VAHC, and STVA - are the independent variables.
Control variables
Following Riahi-Belkaoui (2003) and Zéghal & Maaloul (2010), one of the control variables is the size (SIZE) of the firm, measured by the total value of a firms assets. The other control variable is the amount of leverage (LEV) a firm has, measured by the book value of a firms total assets divided by its book value of common equity within the firm (Lev & Sougiannis, 1996; Zeghal & Maaloul, 2010).
Regression model
A logit regression analysis has been used to investigate the relationship between the backsourcing decision (BD) and the outcome of the VAIC model and its three elements (VACE, VAHC, and STVA). A logit regression analysis is used due to the fact that this study analyses either a “yes or a “no” outcome. The two models are provided in (11) and (12):
BD it = α0+α1VAICit +α2SIZEit +α3 LEV it +εit (11)
Results
Descriptive statistics and correlation analysis
Table 4 gives the descriptive statistics for the variables analysed in this study. The correlations between the VAIC and the individual elements are presented in Table 5, showing that no relationship exists between the VAIC itself and its three constitutive elements. A correlation analysis is of importance because if the individual elements of the VAIC are highly correlated to each other, they are no longer mutually independent. If there would be no independence between the individual elements of VAIC and the VAIC, the outcome of the regression analyses would be almost identical. The Pearson correlations in Table 5 suggest there is no significant relationship between the individual elements of the VAIC model. Tabachnick & Fidell (1996) demonstrate that, in bivariate correlations, a multicollinearity is present only when the correlations between independent variables (in this study, VACE, VAHC, and STVA), are equal to or higher than 0.9. Table 5 demonstrates that the elements are independent from each other and usable for the regression analysis.
Table 4 - Descriptive Statistics
N Minimum Maximum Mean Std.
Deviation VACE 119 -5 18,06 0,9373 2,23191 VAHC 119 -16,1 15,31 0,9806 2,94778 STVA 119 -60,15 13,22 -0,3402 6,07587 VAIC 119 -60,13 19,1 1,5782 7,27423 Valid N (listwise) 119
Table 5 – Correlations between the VAIC and its individual elements (VACE, VAHC, and STVA)
VACE VAHC STVA VAIC
VACE Pearson Correlation 1 VAHC Pearson Correlation 0,044 1 STVA Pearson Correlation 0,041 0,018 1 VAIC Pearson Correlation ,359 * ,433* ,855* 1
Note: *. Correlation is significant at the 0.01 level (2-tailed).
Regression analysis and the IC performance
This study is only looking at the amount of knowledge within a firm in the year preceding the year when the backsourcing announcement was made public. Table 6 presents the values on the firms who made the backsourcing decision. Table 7 presents the figures on the non-backsourcing firms.
Table 6 – IC performance of the Backsourcing Firms Decision to backsource made in year "0"
Years Index value: T = - 1 Y Average value: T = -1 Y compared to - 2 Y in %
N 11 Variables VACE 11 0,39 1% VAHC 11 2,62 9% STVA 11 0,55 4% VAIC 11 3,55 7%
Table 7 – IC performance of the Non-Backsourcing Firms Decision not to backsource made in year
0"
Years Index value: T = - 1Y Average value: T = -1 Y compared to - 2 Y in %
N Variables VACE 108 0,98 6% VAHC 108 0,87 10% STVA 108 -0,42 55% VAIC 108 1,43 17%
It is clear that the backsourcing firms have a higher total VAIC value in the year before their backsourcing announcement. However, the increase in percentages one year before the backsourcing announcement (-1 Y) compared to two years before the backsourcing announcement (-2 Y) shows that the non-backsourcing firms had a relatively higher increase in the VAIC. This implies that non-backsourcing firms have gathered more knowledge compared to the backsourcing firms in the year before the backsourcing announcement. Alternatively, this finding can also relate to the fact that the backsourcing firms already had more knowledge. However, this does not seem to be the case, as, when we compare the VAIC index value of the backsourcing and non-backsourcing firms in Tables 6 and 7, it becomes clear that the non-non-backsourcing firms have a higher VAIC index value.
As the total VAIC index value of the backsourcing firms is higher than that of the non-backsourcing firms, the outcome of the regression analyses on the relationship between the VAIC and the backsourcing announcement does not suggest a significant relationship level (see Table 8). Thus, Hypotheses 1a has to be rejected.
Table 8 – Regression results on VAIC
Therefore, this empirical analysis shows that a relatively higher level of VAIC within a firm does not explain the backsourcing decision of manufacturing firms based in the USA. The results in Table 6 on the individual elements of the VAIC show that VAHC has a rather large impact on the VAIC of the backsourcing firms. The regression analysis in Table 8 shows it has “most” significance of all the variables in the VAIC model, but a significant relationship between VAHC and the backsourcing decision is substantiated by the analyses. Table 9 demonstrates that a higher level of efficiency in HC does not explain the decision of a firm to reallocate its manufacturing facilities from a low-cost region back to the USA. Thus, Hypothesis 2a has to be rejected. Nevertheless, it does seem that the backsourcing firms have more interest in the human capital efficiency compared to the non-backsourcing firms.
Table 9 – Regression results on individual VAIC elements, VACE, VAHC, and STVA Variables in the Equation
B S.E. Wald df Sig. Exp(B)
Step 1a
VAIC 0,053 0,069 0,6 1 0,439* 1,055
Constant -2,41 0,376 41,13 1 0 0,09
Nagelkerke R2 = 0,013
Note: *. Correlation is significant at the 0.01 level
Variables in the Equation
B S.E. Wald df Sig. Exp(B)
Step 1a VACE -0,163 0,239 0,467 1 0,494* 0,849 VAHC 0,137 0,105 1,699 1 0,192* 1,147 STVA 0,06 0,125 0,229 1 0,632* 1,062 Constant -2,37 0,398 35,452 1 0 0,093 Nagelkerke R2 = 0,048
The STVA is not an element of any significance in the relationship with the VAIC itself or with the backsourcing decision. A decision by a manufacturing firm to backsource is therefore not related to the efficient use of the structural capital, which suggests that the last hypothesis, 3a, has to be rejected.
Although the last individual element is not linked to a hypothesis it is still showing that the VACE of the backsourcing firms is lower than the non-backsourcing firms and no significance is found for this relationship.
The Nagelkerke R2 shows that the logit regression analysis for the VAIC has a rather weak fit between the data and the analysis tool, but it is still above the 0.01 significance level. The Nagelkerke R2 for the individual elements of the VAIC has a 0.048 significance level, which indicates that it can explain almost 5% of the difference in the dependent variable or the backsourcing decision.
In sum, the results presented above show no significant effects in the expected areas; therefore, all three hypotheses have to be rejected. Consequently, the more specific sub-hypotheses (expectations) 1b, 2b and 3b are also rejected and the importance of IC, or knowledge, in the backsourcing decision cannot be demonstrated.
Discussion
Backsourcing firms have a higher total VAIC value in the year preceding their backsourcing announcement, and the outcome of the regression analyses on the relationship between the VAIC and the backsourcing announcement does not prove to be significant. The importance of IC, or knowledge, in the backsourcing decision
cannot be demonstrated. The regression analysis shows that the VAHC has the “most” significance of all the variables in the VAIC model, but again no significant relationship between VAHC and the backsourcing decision is found. Neither is the STVA an element of any significance in the relationship with the VAIC and the backsourcing decision.
By looking at the main findings from previous relevant studies, a strong significant relation on the VAHC was at least expected. This is a result of a positive relation between the VAHC and the firm performance in a Greek study (Maditinos et al., 2011). In eight Asian economies a relation is found between the VAHC and the financial and stock performance of banks (Young et al., 2009). In South Africa, no relation is found on IC, or the individual elements, and the firm performance (Firer & Williams, 2003). On the other hand, in Taiwan a positive relation between IC, or VAIC, and the financial and stock performance is found, and VAHC was asserted to be a major element contributing to this relationship (Chen et al., 2005). While the literature review suggests the mixed results on different regions of the world in the studies using different performance measurements, the present study adds new insights to the literature on IC and knowledge. Greater knowledge within a manufacturing firm, or IC, does not impact the backsourcing decision, and neither do the individual elements, VAHC and STVA, of IC.
The lack of the association between the VAIC and the individual elements with the strategic backsourcing decision can be explained by external influences, on the one hand, and the possibility that the VAIC model might not be the most adequate analysing tool for the amount of knowledge or intellectual capital in a firm, on the other hand.
The VAIC model has been used in this study because of its simplicity and reliable data input. The results are not significant and the limitations of the model (see Ståhle et al., 2011) could explain them. Ståhle et al. (2011) studied the elements of the VAIC model and no correlation was found between the firm performance and the VAIC or its individual elements. With regard to the individual elements where no correlation was found, Ståhle et al. (2011) could not offer an explanation because the researchers found elements of the model not clear and not transparent enough. The explanations Ståhle et al. (2011) presents in his paper could have impacted the results of this study. Proof that the explanations given by Ståhle et al. (2011) are applicable to this study is not available due to the fact that Ståhle has not presented clear explanations. Ståhle et al. (2011) studied the relation between firm performance and the VAIC, therefore his results and explanations cannot be presented as reasons for the insignificance for this study. Many other scholars have used the VAIC model and did not suggest that the VAIC model was incapable of measuring intellectual capital within an organisation.
Looking at other explanations that could have influenced the outcome of this study it can be argued that the outsourcing firms could also have generated the backsourcing decision themselves. The outsourcing firms transferred their manufacturing knowledge to the regions where the outsourced activity was created. If local competitors gained new knowledge through the knowledge transfers with the US firms, the local competitors could be manufacturing the same products. In combination with their local knowledge, the local competitors could have got a competitive advantage over the US firms. Said differently, local competitors can produce the same products, but, through the use of more efficient production
processes, they can produce at a lower cost. This can lead to an intensified level of knowledge within the firms of local competitors as compared to the level of knowledge within the American backsourcing firms. Other external influences, such as governmental regulations favouring local producers, could privilege the local organisations as well. This could motivate the US firms to leave such regions and backsource their activities to the USA (Peng, 2012).
This paper studies the impact of the sum of all knowledge that could generate a competitive advantage for the firm. This knowledge consists of tactic and implicit knowledge and the possibility exists that the tactic knowledge is influencing the backsourcing decision more then this study is able to present. Examples of external influences are culture differences and the educational system in the local region. This influence cannot be monitored in this study, which does not mean the influence is not there. The different kinds of knowledge are not used in the analysis of this study and could bring new insights on whether or not a greater knowledge level influences the backsourcing decision.
The RBV was used as a starting point for this study. It posits that intangible assets are the most valuable assets a firm can possess in order to create a sustainable competitive advantage. In this study, the knowledge and intellectual capital of a firm have been considered the most important factor in making the strategic backsourcing decision. While the results show that there is an association, the results did not reach the expected significance levels. This allows us to conclude that the RBV and the knowledge within manufacturing firms do not provide a good foundation for the explanation of the strategic decision to backsource.
Of relevance to the management, our results suggest that knowledge is not a driver in the strategic decision for or against backsourcing. The VAHC was the element in this study expected to yield a significant result due to the fact that if a firm hired more employees, the firm would possess more knowledge. The insignificance of the VAHC in this study indicates that, for manufacturing firms based in the US, a higher human capital level within a firm does not implicate that a firm will choose to backsource its activities to the US. Intellectual capital is an important part of a firm and should be managed, controlled, and invested in through the development of strategies in order to develop the existing knowledge for future opportunities and to establish a better competitive advantage (Kong & Thomson, 2009).
To resolve the issues addressed in this study for future IC or knowledge opportunities, more firms need to be analysed in multiple regions of the world. The VAIC model in combination with a questionnaire could enhance our understanding of the knowledge level in firms. It should also examine if a different relation exist between tactic or explicit knowledge and backsourcing, and implicit knowledge and backsourcing. The VAIC model would then provide robust quantitative results, while questionnaires would generate a deeper qualitative insight. If the backsourcing trend continues to flourish, and if the amount of firms backsourcing their manufacturing activities increases further, analyses similar to the one presented in the present study can be undertaken in the future. Future studies including questionnaire and interview based databases, as used in Germany by Kinkel (2012), could bring new and better insights on the international backsourcing activities of manufacturing firms. The qualitative analysis could then examine, through the use of an adequate questionnaires
design, if the possible internal and external influences, as mentioned in this study, are indeed of influence on the backsoucing decision of firms.
The mixed results in the previous studies suggest the conclusion that knowledge in firms is evolving and varies in different regions in the world. While, as suggested by Kinkel (2012), backsourcing has become a phenomenon, it is however possible that this phenomenon takes different forms in different regions of the world, and generates different outcomes.
Conclusion
This study investigated the possibility of a relation between the intensification of knowledge in a manufacturing firm and the strategic decision to backsource their manufacturing activities back to the USA. A total of 119 firms from the S&P 500 was analysed through the methodology adopted from a series of comparable previous studies (Chen et al., 2004; Chen et al., 2005; Firer & Williams, 2003). IC and two individual elements - the VAHC and STVA - were used for the hypotheses generation. Knowledge could be measured through IC due to the fact that its definition covers the knowledge within a firm quite well. Some practitioners even argue that IC is nothing more than a term that economists and accountants use to refer to knowledge within a firm (Denning, 2009). IC gives the economists and accountants the opportunity to make a differentiation between IC, which is knowledge, and information. In practice, the distinction between knowledge and IC is blurred, as the two can be considered to be the two sides of the same coin.
The VAIC model introduced by Pulic (1998) was used to analyse the level of knowledge within a firm because of its simplicity and reliable data input. Intellectual capital is seen an important part of a firm and should be managed, controlled, and
invested in through the development of strategies. This would entail development of the existing knowledge for future opportunities and a better competitive advantage (Kong & Thomson, 2009). Contrary to the fact that intellectual capital and knowledge are seen as a competitive advantage and strategic assets, this research has generated mixed results and cannot fully substantiate this statement (Peng, 2012).
Backsourcing firms have a higher total VAIC value in the year preceding the backsourcing announcement, but the outcome of the regression analyses on the relationship between the VAIC and the backsourcing announcement does not prove to be significant.Thus,the importance of IC, or knowledge, in the backsourcing decision could not be demonstrated. The regression analysis did show that the VAHC generates “most” significance of all the variables in the VAIC model, and no significant relationship between the VAHC and the backsourcing decision was found. Neither is the STVA an element that impacts the relationship between the VAIC and the backsourcing decision.
Looking at different external influences, the backsourcing decision could also be influenced by other factors, the outsourcing firms themselves transferring their manufacturing knowledge to the regions where the outsourced activity was created. This increases the sum of all knowledge to generate a competitive advantage in a local manufacturing firm leading to a weaker position for the US manufacturing organisation. Other potential factors include culture differences, educational system in the local region, and recruitment activities by local firms, all of which can influence the level of tactic or implicit knowledge within the US organisation or local firms. These elements are important for further research on the underlying, knowledge related, characteristics leading to the backsourcing decision.
