GTID and Master_Delay causes excessive initial delay

When attempting to delay a Slave attached with GTID, there appears to be an extra delay applied initially. For example, this output reflects a Slave that is already delayed by 43200 seconds. When switching to GTID replication, replication is paused until SQL_Remaining_Delay counts down to 0:

MariaDB [(none)]> stop slave; CHANGE MASTER TO master_use_gtid=current_pos; CHANGE MASTER TO MASTER_DELAY=43200; start slave;

Relay_Master_Log_File: bin-log.018039
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Exec_Master_Log_Pos: 625
Seconds_Behind_Master: 44847
Using_Gtid: Current_Pos
SQL_Delay: 43200
SQL_Remaining_Delay: 43089
Slave_SQL_Running_State: Waiting until MASTER_DELAY seconds after master executed event

When switching to conventional replication coordinates, the delay is applied as expected:

MariaDB [(none)]> change master to master_log_file='bin-log.018040', master_log_pos=20289860;
Query OK, 0 rows affected (0.25 sec)

Relay_Master_Log_File: bin-log.018040
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Exec_Master_Log_Pos: 28182467
Seconds_Behind_Master: 43636
Using_Gtid: No
SQL_Delay: 43200
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Unlocking tables