Delta table is the non-hive compatible format. So there must also be permissions for a client to access the path to the database’s location so that it can create a new temporary “directory” there. This comes from Spark SQL’s handling of external tables. Delta tables just happen to always go through this path. 

See Apache Spark code https://github.com/apache/spark/blob/branch-2.4/sql/hive/src/main/scala/org/apache/spark/sql/hive/Hi... for more information