A dynamic model of offshore software development

摘要

As the offshoring of knowledge work has accelerated, theoretical models to explain the phenomenon have not kept up. Most theoretical models assume a static transactional relationship from various factors to a binary offshoring decision. Such models do not take into account the mix of sourcing choices at the level of a firm, nor do they consider dynamic changes over time. To help fill these gaps, we use five case studies on offshore migration of software work by major US companies. Data were collected from senior executives. We use these data to develop a dynamic conceptual model that incorporates three factor groupings which collectively help explain offshore sourcing outcomes: (1) economic factors; (2) the nature of the development activity; and (3) managerial capabilities and practices. Importantly, the model includes five feedback loops among sourcing decisions, sourcing mix, and these three factors. Thus, the relationships in the model are not unidirectional, nor static; rather, they are iterative and dynamic, involving feedback loops, learning, and cumulative effects over time. In this dynamic model, the sourcing 'mix,' a continuously changing offshore portfolio, is a key firm-level dependent variable, closer to the economic concept of a 'stock' measure that represents the cumulative effect of sourcing decisions over time. This variable may be measured in different ways, for instance as the amount of work done offshore, or the number of workers emploved offshore.