This thesis puts a theoretical explanation of governance mode choice developed by Bach and Galvin (2008) to a first empirical test. The theoretical model suggests four alternative governance modes, namely buy, concurrent sourcing, make, and dual distribution. All these modes were found in the empirical setting of the German white wine industry. This specific empirical setting with a focus on the activities harvesting, wine-making, and bottling was chosen in order to avoid biased results because of unknown volume or frequency effects. Furthermore, this industry is not dominated by a few powerful economic actors; governance mode choice is not affected by market power. Finally, as a prerequisit for the validity of the theoretical model, for each of the three value chain activities studied an intermediate market does exist. The empirical results support the theoretical model in most parts. However, some empirical facts gave rise to new questions that are not yet covered in the theoretical rationale. The dissertation ends with suggestions for future research, and theoretical and practical implications.