I wouldn't assume that either. I'm not sure that there's any one metric that does a good job capturing the value of a very good reliever. fWAR isn't sufficient for the reasons I've stated. RA-9 WAR has its own issues, and WPA relies too much on context. I'm comfortable saying that I think Morrow was less valuable than his fWAR indicates, and that I think he would have been a very good reliever, and that I think very good relievers are undervalued by fWAR. I'm not saying it was a mistake that he was used as a starter, but I think it's possible that he would have added more value in relief (probable was too strong, and relies too much on subjective preferences in evaluating how valuable a pitcher is).
You mentioned Cecil earlier. Two people could go on for pages debating whether or not he would have been more valuable as a starter, and both could make valid and convincing arguments. In the end the decision was fine, and it might at the same time have cut into his value to the team.