In each model specification, the coefficient on the contract indicator is negative and is very different from zero. Note that the coefficients in survival models express changes in probability for an event, i.e. the hazard rate. Here, the event is defined as an agreement that has expired. Therefore, a negative coefficient indicates a decrease in the likelihood that an agreement will be struck down when it is entered into as a contract. The results suggest that contracts take much longer than executive agreements and that the difference in sustainability is neither the result of arbitrary technical agreements nor a by-product of a decision-making process primarily motivated by the Senate seat card. Many scientists are skeptical of the usefulness of two apparently similar political instruments for concluding international agreements, most of which are critical of the treaty because of its alleged inflexibility and insignificance. Of course, criticism of the treaty is not new and dates back at least to the 1940s. Footnote 3 During this period, the debate focused on whether the two instruments could be used interchangeably, with a particular emphasis on the constitutional limits of concluding the executive agreements of Congress through treaties. Over time, however, it became clear that neither the courts nor the Department of Foreign Affairs were very concerned about the delineation of the constitutional limits on the interchangeability of the two engagement mechanisms.

Footnote 4 Take, for example, the former U.S. State Department legal adviser, Harold Koh, who suggests that there are only two reasons why the State Department uses treaties, namely the link to Congress and the «powerful political message» conveyed to the world through the treaty ratification process. With regard to the interchangeability of the instruments, Mr. Koh believes that the «long-term» view is that it is constitutionally permissible to use the executive agreements of Congress instead of the treaty. Note 5 52 Martin estimates that the cost gap is the largest for the difference between executive and exclusive contracts, but it extends its argument to the difference between congressional executive agreements and treaties. See Martin, supra note 15, at 447. Although TIF merges these four reasons, some observers might argue that the expiry of an agreement on the basis of its original terms should be treated differently from other grounds for suppression. Finally, an agreement that expires on an expiry clause does not appear to «break» in a manner comparable to an agreement that ends, for example, with a withdrawal. However, the question of whether the parties contain a termination provision that includes an expiration date in an agreement is endogenous, meaning that it can be determined, at least in part, by the same circumstances that lead to the traditional collapse of an agreement. In the related area of private contracts, studies have shown that a contract with an unreliable trading partner not only increases the likelihood that a contract will be terminated prematurely, but also increases the likelihood that a limited-term termination clause will be included in the contract. Footnote 72 Therefore, both agreements that end due to termination, termination or replacement and those that end in accordance with expiry clauses are a reliability of a party.