But most animals don't leave it intact. They chew through it shortly after birth. You can't really have a tissue that is sturdy enough to survive tension during fetal development and vaginal delivery that then instantly falls apart, so it has to be manually severed after delivery. The vast majority of mammals don't let it stay attached for long at all, because their offspring are pretty mobile immediately after birth. From my reading of some of the random websites that recommend this, apparently it was based on the observations of a single species of higher ape (a chimp I think) that doesn't sever the umbilical cord quickly. But when we have been severing cords as a species for generations and the vast majority of other mammals sever the cord with their teeth, I think the evolutionary biology evidence points towards severing the cord quickly.

Now evolutionary biology isn't a solid basis for medical practice, but we don't really have much scientific data at all to base this on at this point. There have been reports of increased rates of serious infections from the practice, which has face validity with the fact that you're leaving a devascularized piece of tissue attached to the vascular system of neonate with an immature immune system. Outside of infection, there has been some case reports of polycythemia (excessively high red blood cell count) and jaundice in these infants. This makes sense physiologically. While attached to the placenta there is a greater intravascular volume available to the infant, which is the entire basis behind delayed cord cutting. It stands to reason that continuing to allow that extra blood volume to enter the infant would result in polycythemia and jaundice.

I'm not intimately familiar with the foundational literature by which the standard DCC cutoffs of 1 minutes or cessation of umbilical pulsatility were founded upon. There could be a very real argument for saying, should the time be 2 minutes? 5 minutes instead of 1? Or should we at least study it if it hasn't been already?

In summary, we have a piece of dead/dying tissue attached to a physiologically stressed neonate with an immature immune system. Leaving it attached for days is in contradiction to the vast majority of other mammalian labor behaviors, is inconsistent with the majority of human's labor history, and has a clear pathological mechanism by which the commonly reported complications can be easily explained. Without some legitimate evidence to actually support benefits or disprove the risks, I think this practice should be discouraged by healthcare professionals.