This article provides a non-technical overview of important results of the game theoretical literature on the formation and stability of international environmental agreements (IEAs) on transboundary pollution control. It starts out by sketching features of first and second best solutions to the problem of transboundary pollution. It then argues that most actual IEAs can be considered at best as third best solutions. Therefore, three questions are raised: 1) Why is there a difference between actual IEAs and first and second best solutions? 2) Which factors determine this difference? 3) Which measures can help to narrow this difference? This article attempts to answer these questions after giving an informal introduction to coalition models.