Byline: Leonardo Baccini, Andreas Dur, Manfred Elsig Existing research indicates the interrelated nature of different dimensions of the design of international institutions. In particular, it shows the greater flexibility of deep agreements. We argue-and demonstrate empirically-that the positive relationship between depth and flexibility holds for preferential trade agreements (PTAs). But we add two qualifications to the conventional wisdom that depth and flexibility go hand in hand. First, we argue that the positive relationship between depth and flexibility proves weaker for democracies than for nondemocracies. Second, when making deep agreements more flexible, countries also add strings to the use of the additional flexibility provisions. An original data set on the design of 587 PTAs allows us to test our arguments. Both descriptive evidence and multivariate statistics support the theoretical expectations. The findings contribute to the literatures on the design of international institutions and the causes and consequences of PTAs. CAPTION(S): Appendix S1. The politics of trade agreement design operationalizing depth.