One probable reason is that institutions do not want to associate themselves with any publisher that might be fraudulent - as in those that do not even have a true editor, but only a desk with a computer on it, and where the only thing the "editor" does is to count the money, and tick the "Accept" button once the check has arrived.
So the indexation is a way - while not a terribly accurate one - to distinguish lousy science from - at least occasionally - rather good science.
The indexing publishers are establishing criteria for inclusion in their databases, and the quality of the indexed journal is known and also its visibility, and that is the reason for institutional choice of journals. In times of information overload, this selection process in important.
Well, not only at Scopus. In the case of Colombia in a wider group of databases. But, the most important for the national agency of Science (Colciencias), are Web of Sciences and Scopus, and among them, according the quartile position, Q1, Q2, Q3 and Q4.
The best catalog and above scopus is journal citation reports (JCR) of thomson reuters, today clarivate analytics, where in addition to being journals with impact factor or to be about to reach it, it has the main journals that many times are not in scopus, so the latter is not the panacea