For the past thirty plus years, firms in various industries, including bio-pharmaceuticals, have moved beyond their organizational and geographic boundaries in search of complementary resources and capabilities, as well as more effective ways to conduct their value-chain activities (Manning et al. 2008; Narula 2001). An impressive number of studies have examined the advantages and disadvantages of firms' boundary-spanning decisions through particular sourcing choices, such as offshoring and outsourcing (domestic or offshore) (Nieto and Rodriguez 2011; Cesarani 2014; Gooris and Peeters 2014; Oshri et al. 2015). However, the majority of these studies considered the above-mentioned sourcing choices independently from each other. Yet, researchers and managers are increasingly recognizing that by considering several sourcing choices simultaneously, firms are able to diversify their options and reduce the drawbacks of relying too heavily on one particular sourcing choice (Contractor et al. 2010; Mudambi and Venzin 2010).
In the present study, we develop the concept of the sourcing portfolio, which we define as the firm's collection of different sourcing choices used to conduct a particular value-chain activity. As identified in prior literature, we focus on four sourcing choices: domestic in-house, domestic outsourcing, offshore outsourcing and captive offshoring (Contractor et al. 2010). Domestic in-house has been the traditional sourcing choice for firms where activity is conducted within the firm's organizational and geographic boundaries. Domestic outsourcing refers to the use of an external domestic vendor to conduct the activity, while offshore outsourcing involves the use of an external foreign vendor. In the case of both domestic and offshore outsourcing, the firm conducts the activity outside the organizational boundaries. When the firm uses its foreign subsidiaries to conduct the activity, meaning that the activity takes place outside of the geographic boundaries of the firm but remains within its organizational boundaries, it is engaged in captive offshoring. In the present research, we examine the configuration of sourcing portfolios based on these four sourcing choices.
The configuration of the sourcing portfolio can be described in terms of portfolio scope--the number of sourcing choices used by a firm at a given time--and portfolio diversity, which reflects the extent to which a firm relies (equally or not) on various sourcing choices (Bruyaka and Durand 2012; Morgan and Rego 2009). Portfolio scope reflects a firm's experience with various alternate sourcing choices, while portfolio diversity is associated with the distribution of resources across various sourcing choices. The diversity of a sourcing portfolio tends to increase when a firm expands the scope of its portfolio, but this may not always be the case. For example, firms with a broad scope for their sourcing portfolio may rely heavily on just one, or a limited number of sourcing choices, thus providing relatively low diversity despite the wide scope. Therefore, it is essential to consider both portfolio scope and diversity when trying to understand a sourcing portfolio configuration.
As firms increasingly adopt a portfolio approach towards their sourcing decisions, we argue that sourcing portfolios are an important unit of analysis. Specifically, we examine two research questions related to sourcing portfolios: (1) how do firms combine different alternate sourcing choices in their sourcing portfolios? and (2) what triggers the evolution of their sourcing portfolio configurations over time? In order to address these research questions, we adopt a co-evolutionary perspective (Lewin and Volberda 1999, 2011), a promising theoretical approach that, to our knowledge, has not yet been applied to study the evolution of portfolio configurations of sourcing decisions.
We study the configuration and evolution of sourcing portfolios in the context of a high value creating activity--drug development (i.e., clinical trials), which is part of the R&D activities of bio-pharmaceutical firms. Drug development is an important and economically significant activity for these firms, as evidenced by the significant portion of total R&D expenditures going towards drug development (PhRMA 2013). At any given time, a firm has multiple ongoing clinical trials, for which a sourcing choice has to be made. In addition, research and development often warrants a portfolio approach, as it helps minimize the high degree of risk and uncertainty associated with these activities (Wuyts and Dutta 2014), which is particularly true for bio-pharmaceutical R&D (McGrath and Nerkar 2004). Thus, drug development activities of bio-pharmaceutical firms constitute an ideal setting for research focused on the evolution of sourcing portfolio configurations.
The co-evolutionary perspective that we adopt to address our research questions calls for a dynamic (i.e., longitudinal), multi-level research design. We conducted a cross-case longitudinal analysis of sourcing portfolios for clinical trials of four firms--two pharmaceutical and two biotechnology firms from North America and Europe (1980s-2010). We use a case study approach instead of quantitative analysis because of our preference for the depth of study over scope. Following the inductive research methodology for case studies (Welch et al. 2010), we (a) track the changing configurations of sourcing portfolios for clinical trials (portfolio scope and diversity), (b) identify a distinct pattern of evolution of sourcing portfolio configurations for clinical trials across geographic and organizational boundaries, and (c) describe key variation, selection and retention (VSR) processes, both in the internal environment (at the portfolio and organizational level) and in the external environment (at the industry and institutional level).
Our study makes the following key contributions to the literature. First, by studying how firms combine different alternate sourcing choices (our first research question), we demonstrate the importance of sourcing portfolios as a unit of analysis for scholarly work on offshoring and outsourcing in general, and of R&D activities in particular. The existing literature on sourcing choices has primarily looked at each sourcing option separately, or examined bi-modal choices (Cantwell 1995; Dunning and Lundan 2009; Martinez-Noya et al. 2013; Wuyts and Dutta 2014). However, a handful of studies--e.g. Contractor et al. (2010) and Mudambi and Venzin (2010)--have stressed the need to combine the literature on offshoring and outsourcing, as firms today have to simultaneously consider all four sourcing options to balance their advantages and disadvantages (van de Vrande 2013). Thus, by considering sourcing portfolios in which various sourcing choices co-exist, our paper addresses an important gap in the literature.
Second, by adopting a co-evolutionary perspective, we describe the triggers of changes in sourcing portfolio configurations (our second research question), thus contributing to the growing research on co-evolutionary firm dynamics (Lewin and Volberda 2011). Specifically, we conceptualize the phenomena of offshoring and outsourcing in a new light (e.g., refining the organizational level of analysis, tracking dynamic changes in internal and external drivers of portfolio configurations evolution). Thus, our study addresses the call by scholars to extend the co-evolutionary logic within organizations (Ingram and Roberts 1999) and to integrate co-evolutionary models with "more tangible notions and observable factors, to further refine it in order to establish closer connections between ideas and to present more coherent and workable theory" (Durand 2006, p. 139). The longitudinal nature of our study allowed us to contribute new evidence on temporal sequence and temporal distribution within a portfolio of firms' key sourcing choices. Specifically, we established the recursive nature of sourcing decisions by providing novel empirical evidence on internal and external triggers of backsourcing, a growing phenomenon whose prevalence, antecedents and outcomes have been under-researched so far.
We proceed by reviewing existing literature on the offshoring and outsourcing of R&D activities, followed by the review of existing studies on offshoring and outsourcing that have adopted a co-evolutionary lens. This concise review enables us to discuss two major research opportunities we address in the present paper--the relevance of considering a sourcing portfolio as a unit of analysis and the relevance of the co-evolutionary perspective. We then introduce the methods of the qualitative case analysis we used in our empirical study and present our findings. We conclude our paper with a discussion of its contributions, limitations, implications for research and academic and managerial practice.
2 Theoretical Grounding
2.1 Offshoring and Outsourcing of R&D
One of the areas in which the importance of outsourcing/offshoring has risen in recent years is R&D related activities (Martinez-Noya and Garcia-Canal 2011). The traditionally preferred choice for R&D activities has been domestic in-house, but the shift to alternate sourcing choices (i.e., outsourcing, offshoring and combinations of both) has been driven by many advantages accruing to firms. Important arguments for offshoring R&D include lower wages and salaries (efficiency seeking motive) (Woodward and Rolf 1993), access to foreign talent and foreign knowledge clusters (knowledge seeking) (Alcacer and Chung 2007; Cantwell and Mudambi 2005), and developing markets abroad (market seeking) (Lewin et al. 2009; Manning et al. 2008; Nieto and Rodriguez 2011). However, the decision to offshore R&D activities entails higher barriers and costs compared to conducting the same task in the home nation of the firm (Hitt et al. 1997). In addition, offshoring can...