A Comparison of Transaction Cost, Agency, and Knowledge-Based Predictors of IT Outsourcing Decisions: A U.S.-Japan Cross-Cultural Field Study

Abstract As outsourcing evolves into a competitive necessity, managers must increasingly contend with the decision about which software development projects to outsource. Although a variety of theories have been invoked to study the initial outsourcing decision, much of this work has relied in isolation on one theoretical perspective. Therefore, the relative importance ascribed by managers to the factors from these theories is poorly understood. The majority of this work also masks interesting insights into outsourcing decisions by focusing on the information technology (IT) function rather than individual projects as the unit of analysis, where many of these decisions occur. In contrast, prior research at the project level has focused on predicting development performance in the postoutsourcing-decision phases of projects. The objective of this study is to examine the relative importance that IT managers ascribe to various factors from three complementary theories—transaction cost economics, agency theory, and knowledge-based theory—as they simultaneously consider them in their project outsourcing decisions. A secondary objective is to assess the cross-cultural robustness (United States versus Japan in this study) of such models in predicting project-level IT outsourcing decisions. We develop and test a multitheoretic model using data on 1,008 project-level decisions collected from 33 Japanese and 55 U.S. managers. Overall, our results provide novel insights into the relative importance that managers ascribe to the factors from these three theories, their complementarities and occasional contradictions, and offer new insights into the differences among U.S. and Japanese IT managers. Implications for theory and practice are also discussed. Keywords: agency theoryconjoint studyIT sourcingJapanese softwareknowledge-based theoryknowledge managementoutsourcingsubcontractingtransaction cost economicsvendor selection