Finally, some scientists have argued that a key difference between congressional contracts and executive agreements lies in the information produced in the process of obtaining legislative approval. Footnote 58 That is, in entering into a contract agreement, the executive must disclose important private information to be persuasive and to ensure the approval of a qualified majority in the Senate. This dynamic can be highlighted by the example of John Yoo, who considers a potential military conflict between the United States and China over territory and negotiations on the division of that territory. The domestic political struggle for treaty approval requires negotiators to communicate to the Senate the chances of a war against China. Yoo argues that observing this process would allow China to obtain more accurate information about the American faith than the congressional executive agreement provides. China can therefore insist that the agreement be concluded in the form of a treaty and, since the underlying information is more precise, it is encouraged to strengthen confidence in continued compliance with the agreement. The implementation of executive agreements increased considerably after 1939. Prior to 1940, the U.S. Senate had ratified 800 treaties and presidents had concluded 1,200 executive agreements; From 1940 to 1989, during World War II and the Cold War, presidents signed nearly 800 treaties, but concluded more than 13,000 executive treaties. Since each observation in the data set is a contractual year, the analysis takes into account differences in the shelf life of different types of agreements, while maintaining consistent characteristics. Differences in shelf life are estimated from the survival time analysis. These methods are also called event history studies. Footnote 88 Before proceeding, it is useful to define a few keywords.
Survival time analysis is mainly used in medical sciences using the most common terminology in clinical trials. A “subject” is an observation unit, here an agreement. An “event,” “death” or “failure” are synonymous with the appearance of the incident of interest, here the expiration of an agreement. “Survival” is the period between the beginning of observation and the occurrence of the incident, here the period during which an agreement is in effect. Agreements in force during the last observation period are considered “legislated,” i.e. with a survival period that has an unknown floor and ceiling, because the final duration of the agreement cannot be respected. Finally, footnote 89, a “risk rate” refers to the probability of an event occurring. 41 id. to 1336 (“Overall, the president will probably have more difficulty withdrawing unilaterally from an agreement between Congress and the executive branch than a Treaty under Article II”).” Some scholars doubt the assertion that presidents can opt out of contracts more easily than agreements between Congress and the executive branch. As Koremenos and Galbraith point out, many agreements in the UN treaty collection have opt-out clauses that would allow a president to legally leave an agreement, regardless of the form in which it was concluded.
See Barbara Koremenos, The Continent of International Law: Explaining Agreement Design 124 (2016) (finding that 70% of agreements have withdrawal provisions based on a sample of contract samples in the UN Treaty Collection); Galbraith, supra note 26, at 1720 (arguing that the withdrawal provisions give the successors of the current president a simple opportunity to legally withdraw from a treaty). For doctrinal challenges to the assertion that contracts can be withdrawn more easily, see Bradley, note 36 above. At the same time, anecdotal evidence also provides alternative explanations for plausibility. Consider, for example, the negotiation process for arms reduction agreements between the United States and Russia.