The Case Zablocki Act of 1972 requires the president to notify the Senate of any executive agreement within 60 days. The Powers of the President to conclude such agreements have not been divided. The notification requirement allowed Congress to vote to cancel an executive agreement or refuse to fund its implementation.   Statistical significance implies no relevance in terms of content. With a large number of observations, as in this study, it is important to supplement the statistical results with evidence that the results are significant in their content and not only marginal. Differences in survival times can be expressed as risk ratios, which describe the risk rate ratio for different subgroups. Here, the risk ratio for the preferred model contract indicator (5) is 0.3, indicating that the relative probability that a contract will expire at any time during the observation period is approximately 30% of the probability that an executive agreement will expire. It is possible to understand this ambiguity in the empirical results by observing that previous studies all take a similar approach. The researcher analyzes the environment in which an international agreement was reached and tries to identify patterns that help predict the type of instrument used.
If the scheme coincides with a motivation that gives a different meaning to treaties and agreements between Congress and the executive branch, it is interpreted as evidence that these instruments differ in quality. For example, based on the conclusion that contracts are used when the partner country has a high GDP per capita, Martin concludes that contracts are preferred when the stakes are high and must therefore be more reliable means of engagement than agreements between Congress and the executive branch. Similarly, Hathaway concludes from the conclusion that few trade-related treaties are concluded that the decision to use the treaty must be inspired by historical conventions that have made the Congress-Executive Agreement attractive for trade negotiations. 104 However, it should be borne in mind that the inclusion of solid effects of the object does not entail major changes in the coefficients. For more differentiated selection effects to fully explain the results, it would have to be assumed that the selection effects within the category are stronger than the selection effects between the categories, for example, that the average difference between two tax treaties explains more differences than the average difference between a tax treaty and an arms control treaty. 103 Formally, it is necessary that the interest cooperative, in this case the indicator of the contract, is not correlated with the potential result, in this case the sustainability of an agreement, after taking into account all the control variables. This is also called the “selection on observables” hypothesis. If the choice between treaties and agreements between Congress and the executive branch were random, the covariate of interest by definition would not correlate with the possible outcome. Without randomization, however, the selection of observable hypotheses remains unverifiable and is the subject of theoretical debate.
An executive agreement is an agreement between the heads of government of two or more countries that has not been ratified by the legislator when the treaties are ratified. .