Global outsourcing is the practice of sourcing from independent suppliers in the global market for goods and services across geopolitical boundaries. As such global outsourcing is part of offshoring, which also includes captive modes of sourcing through the internal organization of the multinational enterprise. Some scholars observed some confusion in the literature regarding the use of these labels (e.g. Bunyaratavej, Hahn, & Doh, 2008; Mol, van Tulder, & Beije, 2005). However, the nature of outsourcing through the market can be fundamentally different from sourcing through internal organization (e.g. Buckley, 2011; Hennart, 2009; Kogut & Zander, 1993). In this study, we focus specifically on global outsourcing defined as the buying by one firm in one country from another independent firm in another country.

Global outsourcing is often considered a critical element of lowcost strategies (Petersen, Prayer, & Scannell, 2000), in gaining access to innovative high-tech inputs (Li, Liu, Li, & Wu, 2008; Linder, 2004), advanced services (Bunyaratavej, Hahn, & Doh, 2007; Kshetri, 2007), global human resources and talent (Lewin, Massini, & Peeters, 2009), and software applications (Verwaal, Commandeur, & Verbeke, 2008). Thus, global outsourcing may be an effective strategy to reduce costs and gain access to worldwide knowledge (Di Gregorio, Musteen, & Thomas, 2009; Li et al., 2008). However, research on the relationship between global outsourcing and firm performance is rare and reports conflicting results (e.g. Jiang, Belohlav, & Young, 2007; Lampel and Bhalla, 2011; Mol, van Tulder, & Beije, 2005). Jiang et al. (2007) estimate the impact of global outsourcing on the market value of Japanese firms. They estimated one model for domestic outsourcing and one model for global outsourcing. They find that global outsourcing has a positive and significant effect on market value, whereas domestic outsourcing has a negative effect. Mol et al. (2005) examined the impact of global outsourcing and firm performance measured as a composite measure of return on sales, return on assets, market share and sales growth relative to the largest competitors. They report that global outsourcing for both measures had no significant impact on firm financial performance. Lampel and Bhalla (2011) find that both benefits and risks increase when high value activities are outsourced, and suggest that the degree to which knowledge is codified and embedded in an activity or location constitutes a barrier to the effective global mobility of knowledge. Furthermore, an increasing number of studies report concerns with respect to global outsourcing on quality (Gray, Tomlin, & Roth, 2009; Steven, Dong, & Corsi, 2014), knowledge and intellectual property rights protection (Roy & Sivakumar, 2011), particularly for high-added value products and services (Lampel & Bhalla, 2011). However, fewstudies investigated the role of innovation in the relationship between global outsourcing and firm financial performance.

Scholars have argued that companies’ choices to engage in international business might positively impact their innovation (e.g. Sapienza, Autio, George, & Zahra, 2006; Siedschlag & Zhang, 2015). Innovation can be directed towards the exploration of new knowledge and skills or the exploitation of existing knowledge and skills (March, 1991). This distinction is important because explorative innovation depends on a higher level of social interaction (Mesquita, Anand, & Brush, 2008), rich (face-to-face) communication (Espinosa, Slaughter, Kraut, & Herbsleb, 2007; Jean, Sinkovics, & Hiebaum, 2014; Mom, van Neerijnen, Reinmoeller, & Verwaal, 2015), and it is more vulnerable to opportunistic behavior, particularly in the absence of proper institutions that protect knowledge and intellectual property rights (Jean et al., 2014). These conditions are often more difficult to realize in global outsourcing relationships. This raises concerns about the impact of global outsourcing on the explorative capacity of the outsourcing firm, and its capacity to generate financial rents from its knowledge assets (Buckley & Strange, 2011).